Abstract
Background
Sustained virological response (SVR12) rates at 12 weeks after treatment for HCV-infected patients with decompensated cirrhosis are used when referring to those with moderate functional impairment, while few data are available for those with more severe impairment. The use of the cirrhosis staging system proposed by D’Amico might provide new insights on timing for antiviral therapy.
Methods
We investigated efficacy (SVR12), safety, and post-treatment variations in clinical and laboratory parameters in 2612 patients with advanced fibrosis (n = 575) or cirrhosis (n = 2037). Cirrhosis was in the compensated phase (without/with varices) or had previously been in the decompensated stage. Different direct-acting antiviral (DAA) regimens were administered in accordance with scientific guidelines.
Results
The SVR12 rate was 97.6% in patients with advanced fibrosis. For patients with cirrhosis, the rate was 96.5% in stage 1, 95.1% in stage 2, 100% in stage 3, 95.7% in stage 4, and 93.6% in stage 5. These rates were independent of gender, age, HCV genotype, and treatment schedule. Positive changes in biochemical parameters and CPT classes following therapy were evident in compensated and previously decompensated patients.
Conclusion
Our findings support the use of DAAs in patients with advanced cirrhosis (stages 3–5) who are at greatest risk and have the most to gain from therapy.
Keywords
1. Introduction
All oral, direct-acting antivirals (DAAs) to treat HCV infection have a dramatic impact on liver cirrhosis, as viral eradication has the potential to improve hepatic function and the patient’s prognosis [
1
, 2
, 3
]. Patients with compensated or previously decompensated cirrhosis treated with DAAs showed rates of sustained virological response (SVR) comparable to those in patients without cirrhosis [4
, 5
, 6
].- Bourlière M.
- Bronowicki J.P.
- de Ledinghen V.
- et al.
Ledispavir-sofosbuvir with or without ribavirin to treat patients with HCV genotype 1 infection and cirrhosis non-responsive to previous protease-inhibitor therapy: a randomized, double-blind, phase 2 trial (SIRIUS).
Lancet Infectious Diseases. 2015; 15: 397-404
The Child–Pugh–Turcotte (CPT) classes are currently used to indicate functional impairment of the liver. However, they do not provide information on other liver disease manifestations with multiple outcome events, each of which may affect the timing of another event developing [
[7]
]. For instance, a patient with cirrhosis staged in CPT class B because of ascites may be re-staged in class A in the event of successful treatment with diuretics, or re-staged in class C if acute-on-chronic liver failure occurs. A new framework for classifying cirrhosis, which incorporates hemodynamic and clinical signs, has been proposed by D’Amico et al. [[8]
]. The authors used two landmark events during liver cirrhosis (the development of esophageal varices and the occurrence of decompensating events) to identify five prognostic stages with a significantly increased risk of mortality. In the D’Amico et al. system [[8]
], portal hypertension rather than impaired synthetic liver function parameters is used to determine patient prognosis. This stratification allows a more precise definition of ‘decompensated’ cirrhosis and provides a prognostic system independent of CPT class, the Model for End-stage Liver Disease (MELD), and the presence of co-morbidities [[8]
]. There is a pressing need to link these stages to clinical outcome, such as long-term outcome after a successful course of antiviral therapy. An initial step toward this approach has been provided by Di Marco et al. [[9]
] who found that in patients with compensated cirrhosis, the SVR rates following peg-interferon and ribavirin treatment differed with stage of portal hypertension: SVR rates were 30.7% and 18.1% in patients without (stage 1) and with (stage 2) esophageal varices, respectively [[10]
]. Data on patients with cirrhosis and previous hepatic decompensation who have undergone treatment with the new DAAs are fragmentary and limited. Guarino et al. found that the SVR rates were lower than those seen in patients with compensated cirrhosis, and ranged from 66.7% to 84.7% [[10]
]. The impact of new DAA administration in patients stratified according to the D’Amico system [[8]
] has not yet been evaluated.Here we report the findings of a prospective, observational cohort study of treated patients with severe liver disease associated with all viral genotypes. We describe the efficacy and safety of new antiviral therapies with every oral DAA regimen in patients with liver cirrhosis, redefined in relation to portal hypertension and the occurrence of previous episodes of decompensation, as indicated by the D’Amico staging system [
[8]
].2. Material and methods
2.1 Study design
This observational, prospective study was conducted by a consortium of 25 community and academic Italian centers which collected post-approval treatment data on consecutive patients with HCV infection from May 2015 to December 2016. In March 2015, second generation DAAs received regulatory approval in Italy from the Agenzia Italiana del Farmaco (AIFA) and access to new DAAs for patients with advanced fibrosis or compensated cirrhosis was prioritized. In this study, the prescribing clinician selected the particular treatment regimen according to the viral genotype/subtype, as suggested by the Italian Association for the Study of the Liver [
[11]
], and chose the ribavirin dosage, while treatment duration (12 or 24 weeks) was in accordance with the severity of liver disease. All patients who received at least one dose of the new DAAs were included. Approval to tabulate and analyze the data was obtained from the Ethics Committee of the Coordinating Center in San Giovanni Rotondo. The study was not supported by any pharmaceutical company or other funding agency.Associazione Italiana per lo Studio del Fegato. Documento di indirizzo dell’Associazione Italiana per lo Studio del Fegato per l’uso razionale di antivirali diretti di seconda generazione nelle categorie di pazienti affetti da epatite C cronica ammesse alla rimborsabilità in Italia. Available at: webaisf.org/pubblicazioni/documento-aisf-hcv-2016.aspx. [Accessed 3 March 2017].
Admission criteria were adult age, no evidence of active hepatocellular carcinoma (HCC) on imaging, RNA for HCV identified by PCR, and no previous similar treatment or failure of a previous course with peg-interferon plus ribavirin in combination or not with first generation DAAs. Patients with cirrhosis were offered treatment whether or not they had previous decompensation controlled before starting DAAs. Co-morbidities were recorded and classified according to Harboun and Ankri [
[12]
]. Similarly, co-administered drugs were listed and interaction with the intended DAAs carefully checked by consulting the Liverpool HEP drug interactions guidance [[13]
]. Exclusion criteria were HIV coinfection, major co-morbidities, an estimated glomerular filtration rate of <15 mL/min, and current hepatic decompensation unresponsive to appropriate therapy.University of Liverpool. HEP drug interactions. Available at: http://www.hep-druginteractions.org. [Accessed 3 March 2017].
The characteristics of enrolled subjects were captured either at baseline or at each study visit and included: gender, age, body weight and height, previous treatment (whether naïve or experienced with peg-interferon/ribavirin and/or first-generation antivirals), cirrhotic status, decompensation events (ascites, esophageal bleeding, encephalopathy, hepatocellular carcinoma), listed or not for an orthotopic liver transplantation (OLT), and co-morbidities [
[12]
]. In addition, the following parameters were noted: albumin, INR, bilirubin, platelet count, creatinine, and serum HCV RNA level and genotypes. Determination of portal hypertension was in accordance with the Baveno VI Consensus Workshop [[14]
], and included esophageal varices detected at screening endoscopy, imaging showing collateral circulation, liver stiffness of ≥20–25 kPa, or a platelet count of <150,000 in patients with liver stiffness of <20 kPa.Patients were reviewed monthly during antiviral treatment and virological response was assessed by quantitative HCV RNA at week 4, at the end of treatment, and at 4 and 12 weeks of follow-up. During the study period, we recorded early discontinuation of therapy or liver decompensation events (ascites, hepatic encephalopathy, jaundice, spontaneous bacterial peritonitis, and variceal hemorrhage), HCC, OLT, or death (liver-related or non-liver related). Adverse events due to therapy were categorized according to the Common Terminology Criteria for Adverse Events (version 4.03) [
[15]
].2.2 Definition and staging of cirrhosis
A centralized team of monitors reviewed all records obtained from the various centers. Patients were categorized as having cirrhosis or not according to (1) whether a cirrhotic liver had been documented histologically at any time before enrollment or, in patients without histology, had been determined by FibroScan, and (2) a combination of clinical, laboratory, and abdominal ultrasound parameters established a priori [
[16]
]. Patients were considered to have cirrhosis if they showed the following: (1) esophageal varices at endoscopy or evidence of cirrhosis and/or portal hypertension and/or ascites at ultrasound; (2) a liver stiffness value on FibroScan of >14.0 kPa plus a platelet count of ≤140,000/μL; or (3) a FibroScan value of 10–14 kPa and a platelet count of ≤100,000/μL. Patients with a liver stiffness value of <14 kPa, no hepatic decompensation, and no ultrasound signs of portal hypertension were considered to have advanced fibrosis (F3).For patients with cirrhosis, CPT classes and MELD scores were centrally calculated using site-derived laboratory parameters. Patients with cirrhosis were further stratified into five classes according to D’Amico et al. [
[8]
], as follows: patients with compensated cirrhosis without or with portal hypertension were included in stage 1 or 2, respectively; those with previously decompensated cirrhosis (patients with a history of variceal bleeding) in stage 3; those with a previous, single episode of ascites or encephalopathy in the absence of an esophageal bleed in stage 4; and those with multiple, recurrent episodes of decompensation in stage 5.2.3 Primary outcomes
The primary endpoint was sustained virological response at 12 weeks (SVR12) after therapy in patients with advanced fibrosis and in patients with cirrhosis staged according to D’Amico et al. [
[8]
]. Secondary outcomes included pre- and post-treatment variations in biochemical parameters and hepatic decompensation events during the study period. The comparative efficacy of the different treatment schedules was also controlled in patients with HCV genotypes (GT) 1A, 1B, or 4.2.3 Data analysis
Before the databases from participating centers were pooled, two independent monitors (A.I. and M.L.) systematically assessed data entries for completeness and consistency, and doubts concerning unlikely values were resolved with additional queries. Separate analyses for SVR12 were performed for each genotype, and for GT1 by subtypes 1A and 1B. Within each genotype, analyses were separated according to the presence or absence of cirrhosis and according to previous treatment with peg-interferon/ribavirin (with or without DAAs). Since no head-to-head randomized trials comparing the efficacy of the different DAAs are available, and there were no sample size constraints in our cohort, supplemental analyses of the comparative efficacy of administered DAAs were also conducted; these analyses were restricted to patients with HCV GT1A, 1B, and 4, as a single regimen is considered optimal for HCV GT2 and 3 [
7
, 8
, 11
]. The relapse rate, treatment completion, and frequency of adverse events during or at 3 months after treatment were calculated for the entire study population and for subpopulations. Baseline characteristics were assessed with standard descriptive statistics. Continuous variables were expressed as median values and compared using the Mann–Whitney U test. Categorical variables were reported as percentages and compared using the χ2 test (or Fisher’s exact test, when needed). Variables with p< 0.05 were included in a multivariate stepwise logistic regression model. In order to compare frequencies at baseline and at the end of follow-up, the McNemar test was used for categorical variables, and the Wilcoxon signed-rank test for continuous variables. Analyses were performed using SPSS software, version 13.0 (SPSS Inc., Chicago, IL, USA).Associazione Italiana per lo Studio del Fegato. Documento di indirizzo dell’Associazione Italiana per lo Studio del Fegato per l’uso razionale di antivirali diretti di seconda generazione nelle categorie di pazienti affetti da epatite C cronica ammesse alla rimborsabilità in Italia. Available at: webaisf.org/pubblicazioni/documento-aisf-hcv-2016.aspx. [Accessed 3 March 2017].
3. Results
A total of 2979 patients met the inclusion criteria. However, 367 were excluded from further analysis; these comprised 14 patients treated with peg-interferon in combination with new DAAs, 140 patients with HCV GT 1, 3, and 4 treated with the sub-optimal schedule of sofosbuvir/ribavirin, and 213 patients still on treatment. The characteristics of the remaining 2612 patients are shown in Table 1. Enrolled patients were adults with a mean age of 64.8 ± 11.0 years, 39.5% were above 70 years of age, and 57.4% were male. The frequency of HCV genotypes was as follows: GT1A was noted in 224 patients (8.6%), GT1 B in 1590 (60.9%), GT2 in 494 (18.9%), GT3 in 157 (6.0%), GT4 in 135 (5.2%), and mixed or untyped genotypes in the remaining 12 patients. Advanced liver fibrosis was seen in 575 patients (22.0%), and cirrhosis in the remaining 2037 (78.0%). The majority of patients with cirrhosis were in CPT class A (86.4%), while 249 were in class B (12.2%) and the remaining 29 (1.4%) in class C. In addition, 96.5% had a MELD score of <16. According to the D’Amico staging system [
[8]
], 1758 had compensated cirrhosis without (stage 1, n = 1172) or with portal hypertension (stage 2, n = 586), while the remaining 279 patients had experienced a decompensation event, such as variceal bleeding (stage 3, n = 22), ascites or encephalopathy (stage 4, n = 210), or multiple events (stage 5, n = 47). A total of 107 patients had previously been treated for HCC, and 89% of these had not had a recurrence for at least 6 months. Regarding previous treatment, 1383 patients were naïve to therapy, while the remaining 1229 (47.1%) had been exposed to peg-interferon/ribavirin therapy either alone or in combination with first-generation DAAs. Treatment duration was 12 weeks for 93% of patients, and 24 weeks for the rest. Except for GT2 patients for whom ribavirin administration was mandatory in conjunction with sofosbuvir, ribavirin use was at the discretion of the prescribing physician, and was given to 1114 patients.Table 1Baseline features and sustained viral clearance in 2612 HCV-infected patients after direct-acting antiviral treatment.
| Sustained virological response | |||
|---|---|---|---|
| Total | N (%) | p Value | |
| Overall | 2612 | 2516 (96) | 0.073 |
| Advanced fibrosis | 575 | 561 (98) | |
| Cirrhosis | 2037 | 1955 (96) | |
| Gender | 0.407 | ||
| Male | 1498 | 1439 (96) | |
| Female | 1114 | 1077 (97) | |
| Age (years) | 0.654 | ||
| Median (IQR) | 67 (58–73) | ||
| <70 | 1581 | 1525 (96) | |
| ≥70 | 1031 | 991 (96) | |
| Genotype | 0.246 | ||
| 1A | 224 | 216 (96) | |
| 1B | 1590 | 1535 (97) | |
| 2 | 494 | 476 (96) | |
| 3 | 157 | 146 (93) | |
| 4 | 135 | 132 (98) | |
| Mixed | 12 | 11 (92) | |
| Albumin (g/dL) | 0.139 | ||
| >3.5 | 2137 | 2065 (97) | |
| 2.8–3.5 | 436 | 415 (95) | |
| <2.8 | 39 | 36 92) | |
| INR | 0.898 | ||
| <1.7 | 2532 | 2439 (96) | |
| 1.7–2.2 | 62 | 60 (97) | |
| >2.2 | 18 | 17 (94) | |
| Bilirubin (mg/dL) | 0.494 | ||
| <2 | 2431 | 2344 (96) | |
| (2–3) | 147 | 139 95) | |
| >3 | 34 | 33 (97) | |
| Glomerular filtration rate | 0.002 | ||
| >15–29 | 8 | 6 (75) | |
| 30–59 | 387 | 368 (95) | |
| 60–89 | 981 | 954 (97) | |
| ≥90 | 1236 | 1188 (96) | |
| Pre-therapy status | 0.555 | ||
| Naïve | 1383 | 1335 (97) | |
| Treatment experienced | 1229 | 1181 (96) | |
| Ribavirin | 0.492 | ||
| Without | 1004 | 970 (97) | |
| With | 1114 | 1070 (96) | |
| Co-morbidities (n) | 0.294 | ||
| 0 | 1181 | 1145 (97) | |
| 1–2 | 1276 | 1222 (96) | |
| ≥3 | 155 | 149 (96) | |
INR, international normalized ratio.
a Genotype 2 excluded.
3.1 Treatment response
At 12-week follow-up after therapy, 2516 patients had an SVR12 (96.3%) and 15 patients had relapsed. The patient and virus characteristics which may have impacted on the outcome of therapy are shown in Table 1: no influence of gender, age, number of comorbidities, adjunctive treatment with ribavirin, or failed previous antiviral treatment was seen. In particular, all HCV genotypes were equally responsive to the given therapy. The occurrence of cirrhosis marginally reduced SVR12 rates (p = 0.073).
3.2 Staging of patients with cirrhosis and SVR12 rates
SVR12 rates for the 2037 patients with cirrhosis were evaluated in relation to CTP class, MELD score, and the D’Amico classification [
[8]
] (Table 2). For CPT classes A–C, the SVR12 rates were 96.2%, 95.2%, and 89.7%, respectively; differences were not significant (p = 0.164). Using the MELD score for stratification, the SVR12 rates were 95.9%, 97.8%, and 95.8% for scores of <12, 12–15, and ≥16 (p = 0.672). When the D’Amico staging system [[8]
] was used, the SVR12 rates were 96.5% in patients without portal hypertension (stage 1), 95.1% in patients with compensated cirrhosis and non-bleeding varices (stage 2), 100% in stage 3 patients, 95.7% in patients with a single, previous decompensation event (stage 4), and 93.6% in patients with multiple decompensated events (stage 5); the difference among the five stages was not significant (p = 0.438). SVR12 rates for the 2037 patients with cirrhosis in relation to baseline characteristics are given in Supplementary Table 1.Table 2Sustained virological responses after direct-acting antiviral treatment in 2037 patients with cirrhosis, staged according to either the D'Amico system or the Child–Pugh–Turcotte (CPT) classification.
| Stage of cirrhosis | Sustained virological response | ||
|---|---|---|---|
| Total | CPT | N (%) | |
| 1 (no portal hypertension) | 1172 | 1131 (97) | |
| A | 1093 (97) | ||
| B | 38 (88) | ||
| 2 (portal hypertension) | 586 | 557 (95) | |
| A | 509 (95) | ||
| B | 48 (100) | ||
| 3 (variceal bleeding) | 22 | 22 (100) | |
| A | 17 (100) | ||
| B | 5 (100) | ||
| 4 (single decompensation event) | 210 | 201 (96) | |
| A | 64 (98) | ||
| B | 119 (94) | ||
| C | 18 (95) | ||
| 5 (multiple decompensation events) | 47 | 44 (94) | |
| A | 9 (90) | ||
| B | 27 (100) | ||
| C | 8 (80) | ||
3.3 Comparative effectiveness of the five treatment schedules
Five DAA regimens were administered, namely sofosbuvir/ribavirin (n = 436), sofosbuvir/simeprevir ± ribavirin (n = 452), sofosbuvir/daclatasvir ± ribavirin (n = 293), sofosbuvir/ledipasvir ± ribavirin (n = 733), and dasabuvir/ombitasvir/paritaprevir/ritonavir ± ribavirin (n = 686). The SVR12 rates by therapeutic schedule and the different HCV genotypes in patients with advanced fibrosis or liver cirrhosis are given in Supplementary Table 2. The recommended regimens for GT2 and GT3, namely sofosbuvir/ribavirin and sofosbuvir/daclatasvir ± ribavirin, were successful in most patients. The three regimens for GT1A, 1B, and 4 (i.e., sofosbuvir/daclatasvir ± ribavirin, sofosbuvir/ledipasvir ± ribavirin, and dasabuvir/ombitasvir/paritaprevir/ritonavir ± ribavirin) produced very similar SVR12 rates.
3.4 Treatment safety (Table 3)
A total of 365 adverse events during treatment were experienced by 209 patients (8.0% of the entire cohort): 168 patients had one or two events, while the remaining 41 patients had multiple events. The majority of events (320 of 365, 87.7%) were of mild/moderate severity (grade 1 or 2), while the remaining 45 were severe or life-threatening. The types of adverse events are shown in Table 3: the most common adverse events were fatigue and pruritus (especially in patients receiving ribavirin). Sixteen patients discontinued treatment because of fatigue and diffuse myalgia (5 cases), severe anemia (3 cases), itching (3 cases), worsening of liver function (2 cases), vertigo (2 cases), or pneumonia (1 case). Adverse events stratified by degree of liver functional impairment are shown in Supplementary Table 3. Only two mild events were seen in CPT class C patients, while only 11 mild events were noted in 279 patients in stages 3–5 of the D’Amico system.
Table 3Types of adverse events seen during antiviral therapy.
| Total events | Mild events (grades 1–2) | Serious events (grades 3–4) | |
|---|---|---|---|
| N (%) | N (%) | N (%) | |
| Fatigue | 97 (26.8) | 94 (29.6) | 3 (6.7) |
| Dizziness/headache | 42 (11.5) | 36 (11.2) | 6 (13.3) |
| Irritability | 41 (11.2) | 39 (12.2) | 2 (4.5) |
| Anemia | 39 (10.7) | 32 (10.0) | 7 (15.6) |
| Gastrointestinal symptoms | 53 (14.5) | 47 (14.6) | 6 (13.3) |
| Rash/pruritus | 61 (16.7) | 51 (15.9) | 10 (22.2) |
| Arthralgia | 18 (4.9) | 14 (4.4) | 4 (8.9) |
| Epistaxis | 2 (0.5) | 2 (0.6) | – |
| Infective events | 6 (1.6) | 1 (0.3) | 5 (11.1) |
| Worsening liver function | 4 (1.1) | 3 (0.9) | 1 (2.2) |
| Worsening kidney function | 2 (0.5) | 1 (0.3) | 1 (2.2) |
| Total | 365 | 320 | 45 |
3.5 Functional outcomes
Biochemical parameters and CPT class following therapy were not available for all patients. In patients with paired pre- and post-treatment data, improvements were evident in patients with compensated (stages 1 and 2) and with previously decompensated cirrhosis (stages 3–5). Pre- and post-therapy data were available for 198 patients in stages 3–5 (Table 4): with the exception of bilirubin levels, the numbers of patients with normal albumin and INR values increased significantly compared to baseline. Circulating platelets and creatinine levels also increased significantly with respect to baseline. Pre- and post-treatment information for CPT class was available for 198 patients in stages 3–5: the frequency of CPT class A patients increased significantly from 35.9% to 80.3% (p< 0.001).
Table 4Changes in biochemical parameters and Child–Pugh–Turcotte classes before and following therapy with DAAs in decompensated cirrhotic patients.
| Stage (1-2-3-5) | Baseline | 3 month follow-up | p Value |
|---|---|---|---|
| Albumin (g/dL) (n = 1881) | <0.001 | ||
| >3.5 | 1572 (83.6) | 1725 (91.7) | |
| 2.8–3.5 | 286 (15.2) | 131 (7.0) | |
| <2.8 | 23 (1.2) | 25 (1.3) | |
| INR (n = 1903) | |||
| ≤1.7 | 1849 (97.2) | 1861 (97.8) | 0.001 |
| 1.7–2.2 | 45 (2.3) | 22 (1.2) | |
| >2.2 | 9 (0.5) | 20 (1.0) | |
| Bilirubin (mg/dL) (n = 2006) | |||
| <2 | 1866 (93.0) | 1899 (94.7) | 0.005 |
| 2–3 | 117 (5.8) | 75 (3.7) | |
| ≥3 | 23 (1.2) | 32 (1.6) | |
| Creatinine (n = 1979) | |||
| Median (IQR) | 0.78 (0.68–0.90) | 0.80 (0.70–0.90) | <0.001 |
| Platelets (n = 2025) | |||
| Median (IQR) | 130 (92–174) | 137 (98–184) | <0.001 |
| CPT, n (%) (n = 1822) | |||
| A | 1644 (90.2) | 1733 (95.1) | <0.001 |
| B | 156 (8.6) | 79 (4.3) | |
| C | 22 (1.2) | 10 (0.6) |
Hepatic decompensation events during and following therapy are shown in Table 5A, Table 5B. Two HCCs (0.35%) developed in the 575 patients with advanced fibrosis, one during and one after therapy. The numbers of patients with cirrhosis who experienced adverse events during treatment increased linearly with progression of the staging class of both the CPT classification and the D’Amico staging system. The same trend was apparent in follow-up after treatment. A total of 111 adverse events were seen in the cohort of 2037 patients with cirrhosis (5.45%); of these, 34 were HCCs and the remaining 77 were hepatic decompensation events (ascites, encephalopathy, and bleeding). Of the 34 HCCs, 11 were diagnosed during and 23 following treatment. The majority of HCCs (24 of 34, 70.6%) occurred patients with compensated cirrhosis (CPT class A, or stage 1 or 2). Both during and after treatment, hepatic failure events were seen more frequently in patients who had experienced a past episode of decompensation: 34 of 48 events (70.8%) in CPT classes B and C, and 27 of 48 events (56.3%) in stages 3–5.
Table 5ADecompensation events in 2612 patients with HCV infection during treatment with direct-acting antiviral agents.
| Liver damage | Total patients (N) | During treatment | Ascites/PBS | EPS | Bleeding | HCC | |
|---|---|---|---|---|---|---|---|
| Patients | Events | ||||||
| N (%) | N | ||||||
| F3 | 575 | – | – | – | – | – | – |
| Stage of cirrhosis | |||||||
| 1 | 1172 | 1 (0.09) | 1 | – | – | – | 1 |
| 2 | 586 | 12 (2.05) | 13 | 3 | 2 | 2 | 6 |
| 3 | 22 | 2 (9.09) | 3 | – | – | 3 | – |
| 4 | 210 | 14 (6.67) | 20 | 15 | 1 | 2 | 2 |
| 5 | 47 | 6 (12.77) | 11 | 3 | 2 | 4 | 2 |
| CPT | |||||||
| A | 1759 | 15 (0.85) | 21 | 6 | – | 7 | 8 |
| B | 249 | 15 (6.02) | 20 | 12 | 3 | 3 | 2 |
| C | 29 | 5 (17.24) | 7 | 3 | 2 | 1 | 1 |
Table 5BDecompensation events in 2601 patients with HCV infection at 3-month follow-up after treatment
| Liver damage | Total patients (N) | At 3-month follow-up | Ascites/PBS | EPS | Bleeding | HCC | |
|---|---|---|---|---|---|---|---|
| Patients | Events no. | ||||||
| N (%) | |||||||
| F3 | 569 | 1 (0.18) | 1 | – | – | – | 1 |
| Stage of cirrhosis | |||||||
| 1 | 1165 | 11 (0.95) | 11 | 5 | – | – | 6 |
| 2 | 586 | 22 (3.81) | 25 | 7 | 2 | 5 | 11 |
| 3 | 22 | – | – | – | – | – | – |
| 4 | 212 | 16 (7.66) | 17 | 11 | 2 | – | 5 |
| 5 | 47 | 9 (19.57) | 10 | 5 | 4 | 1 | 6 |
| CPT | |||||||
| A | 1750 | 33 (1.89) | 36 | 11 | 2 | 5 | 18 |
| B | 253 | 19 (7.72) | 20 | 13 | 3 | 1 | 4 |
| C | 29 | 6 (21.43) | 7 | 4 | 3 | – | 2 |
HCC, hepatocellular carcinoma.
a 11 patients who lost to follow-up were excluded.
4. Discussion
The present study conducted by a network of 25 Italian centers involved in the management of patients with HCV-related liver disease, has reviewed the efficacy and safety of DAA treatment in patients with compensated cirrhosis, particularly those with previous decompensation. Our investigation examined a cohort of patients with previously decompensated cirrhosis (n = 279) and reported an unexpectedly high rate of SVR12 (95.7%), comparable with that of patients with less advanced liver damage.
Information on the therapeutic response of patients with CPT class C cirrhosis to DAAs is fragmentary, as safety concerns regarding DAA use has limited their enrollment in registered trials [
4
, 5
, 6
]. Data are numerically more consistent for CPT class B, but the characteristics of patients in this subgroup are very heterogeneous. Indeed, CPT class C patients with decompensated cirrhosis (ascites, bleeding, encephalopathy, jaundice) may be reclassified into class B following successful treatment of the event. In addition, CPT class B can include both patients with a past decompensated event and minimal alteration in biochemical parameters (INR, bilirubin, albumin) and patients who have never decompensated but have larger changes in hepatic indices. There is a confusing relationship between CPT class and impairment of liver function. CPT classification considers five variables, of which two are clinical events consequent to portal hypertension (ascites and encephalopathy) and three are laboratory tests likely reflecting impaired liver function. Although the laboratory parameters are only marginally affected by therapeutic interventions, ascites and encephalopathy may resolve following treatment but recur during follow-up. This variability means an individual patient may be classified differently over time, with a likely transition from one class to another one during CPT staging. In contrast, the D’Amico staging system [- Bourlière M.
- Bronowicki J.P.
- de Ledinghen V.
- et al.
Ledispavir-sofosbuvir with or without ribavirin to treat patients with HCV genotype 1 infection and cirrhosis non-responsive to previous protease-inhibitor therapy: a randomized, double-blind, phase 2 trial (SIRIUS).
Lancet Infectious Diseases. 2015; 15: 397-404
[8]
] only considers the status of cirrhosis in relation to portal hypertension, the most important consequence of liver fibrosis and a harbinger of episodes of decompensation. Adoption of this system means a patient with cirrhosis and ascites remains in the same stage even after successful treatment of the event. In our investigation, SVR rates ranged from 96.5% in the absence of portal hypertension, to 95.1% in the presence of varices (stage 2), 100% and 95.7% in the event of past decompensation (stages 3 and 4, respectively), and 93.6% in patients with multiple events (stage 5). Our findings suggest the use of DAAs for previously decompensated (stages 3–5) patients urgently needing for a cure of their HCV infection could be expanded.As there are no head-to-head randomized controlled trials comparing DAA regimes, and our data were sufficient to conduct supplemental analyses, we undertook indirect treatment comparisons of the efficacy of the different DAA schedules. The analysis provided comparable SVR rates for all regimens: 95.4% for sofosbuvir/simeprevir ± ribavirin, 97.4% for sofosbuvir/ledipasvir ± ribavirin, and 96.4% for dasabuvir/ombitasvir/paritaprevir/ritonavir ± ribavirin. For countries with budget constraints, our results would allow therapeutic choice to be governed only by the cost of the drug(s).
In general, patients with cirrhosis who received treatment with the new DAAs whether or not in registered trials [
[10]
] had lower rates of SVR compared to patients in the present cohort. A likely explanation may be the exclusion from our analysis of patients treated with sofosbuvir/ribavirin, a regimen considered of limited efficacy for GT1 and GT4 patients with cirrhosis. A further reason could be the meticulous checking for potential drug interactions between the DAAs and other agents co-administered to the patient [[13]
].University of Liverpool. HEP drug interactions. Available at: http://www.hep-druginteractions.org. [Accessed 3 March 2017].
Benefit from therapy in this fragile population should not obscure the safety issue. Weker et al. have recently documented the frequency of hepatic decompensation during the course of DAA treatment for hepatitis C [
[17]
]. Gray et al. have also reported a 6% rate of on-treatment mortality in patients who were in CPT class B at baseline, and of 21% in those in CPT class C [[18]
]. This high frequency of serious adverse events has been reported with virtually all oral regimens [[19]
]. However, the alternative view that these events may be unrelated and coincidental to therapy, representing progression of the disease despite antiviral therapy, has been discussed [[20]
]. The safety profile in our real-world cohort was excellent: adverse events occurred in a few patients but the majority of events were mild to moderate and aggravated by ribavirin co-administration. It should be remembered that a consistent proportion of our patients with cirrhosis received off-label therapy with simeprevir-containing schedules and a dasabuvir/ombitasvir/paritaprevir/ritonavir combination, regimens that are not recommended in CPT B and C cirrhosis [[15]
]. Despite this, only three life-threatening events were seen in the 27 patients in these classes.A favorable outcome following DAA administration has been consistently documented in previous studies [
1
, 2
, 10
]: in general, improvements in CPT classes and/or MELD scores have been noted in about one third of treated patients, and are usually thought to reflect better liver function. However, these two staging systems do not only reflect hepatic synthetic function, as two portal hypertension adverse events are included in the CPT classification, and creatinine value in the MELD scoring. Our data in previously decompensated patients revealed improvements in two liver synthetic indices, in platelets counts, and in creatinine values.In conclusion, the main finding after treating 2612 HCV-infected patients with advanced fibrosis or cirrhosis in a real-world setting is that the treatment is safe and efficacious even for patients with previously decompensated disease. With the recommendation to check for drug interactions between DAAs and other co-administered agents, this subgroup of patients may benefit the most from DAA treatment: following a positive therapeutic outcome, parameters of liver synthetic capability are improved and the number of hepatic decompensation events is reduced.
Conflict of interest
None declared.
Co-authors
Angelo Iacobellis: Division of Gastroenterology, “Casa Sollievo Sofferenza” Hospital, IRCCS, San Giovanni Rotondo, Italy.
Immacolata Carraturo: Division of Infectious Diseases, “V. Fazzi” Hospital, Lecce, Italy.
Pieraldo Paiano: Division of Gastroenterology, Hospital of Scorrano, Scorrano, Italy.
Nicola Andriulli and Marta Librandi: Department of Physiology, Faculty of Pharmacia, “La Sapienza” University, Rome, Italy.
Silvia Martini: Department of Medical Sciences, University of Turin and Department of Gastroenterology and Hepatology, Azienda Ospedaliera Città della Salute e della Scienza, Turin, Italy.
Appendix A. Supplementary data
The following are Supplementary data to this article:
References
- Improvement of liver function parameters in advanced HCV-associated liver cirrhosis by IFN-free antiviral therapies.Alimentary Pharmacology Therapeutics. 2015; 42: 889-901
- Interferon-free regimens for chronic hepatitis C overcome the effects of portal hypertension on virological responses.Alimentary Pharmacology Therapeutics. 2015; 42: 707-718
- Long-term outcome after antiviral therapy of patients with hepatitis C virus infection and decompensated cirrhosis.Clinical Gastroenterology and Hepatology. 2011; 9: 249-253
- Daclatasvir plus sofosbuvir for previously treated or untreated chronic HCV Infection.New England Journal of Medicine. 2014; 370: 211-221
- ABT-450/r-ombitasvir and dasabuvir with ribavirin for hepatitis C cirrhosis.New England Journal of Medicine. 2014; 370: 1973-1982
- Ledispavir-sofosbuvir with or without ribavirin to treat patients with HCV genotype 1 infection and cirrhosis non-responsive to previous protease-inhibitor therapy: a randomized, double-blind, phase 2 trial (SIRIUS).Lancet Infectious Diseases. 2015; 15: 397-404
- The clinical course of cirrhosis: the importance of multistate models and competing risks analysis.Hepatology. 2014; 62: 292-302
- Competing risks and prognostic stages of cirrhosis: a 25-year inception color study of 494 patients.Alimentary Pharmacology Therapeutics. 2014; 39: 1180-1193
- Effects of viral eradication in patients with HCV and cirrhosis differ with stage of portal hypertension.Gastroenterology. 2016; 151: 130-139
- Systematic review: interferon-free regimens for patients with HCV-related Child C cirrhosis.Alimentary Pharmacology Therapeutics. 2017; ([in press])
Associazione Italiana per lo Studio del Fegato. Documento di indirizzo dell’Associazione Italiana per lo Studio del Fegato per l’uso razionale di antivirali diretti di seconda generazione nelle categorie di pazienti affetti da epatite C cronica ammesse alla rimborsabilità in Italia. Available at: webaisf.org/pubblicazioni/documento-aisf-hcv-2016.aspx. [Accessed 3 March 2017].
- Comorbidity indexes: review of the literature and application to studies of elderly population.Revue d Épidémiologie et de Santé Publique. 2001; 49: 287-298
University of Liverpool. HEP drug interactions. Available at: http://www.hep-druginteractions.org. [Accessed 3 March 2017].
- Expanding consensus in portal hypertension. Role of the Baveno VI Consensus Workshop: stratifying risk and individualizing care for portal hypertension.Journal of Hepatology. 2015; 63: 743-752
- Common terminology criteria for adverse events (version 4.03).US Department of Health and Human Services, 2010 (14 June)
- External validation of noninvasive evaluation of liver fibrosis in HCV chronic hepatitis.American Journal of Gastroenterology. 2005; 100: 868-873
- Lactic acidosis in patients with hepatitis C virus related cirrhosis and combined ribavirin/sofosbuvir treatment.Journal of Hepatology. 2016; 64: 790-799
- High mortality during direct acting antiviral therapy for hepatitis C patients with Child C cirrhosis: results of the Irish Early Access Programme.Journal of Hepatology. 2016; 65: 446-448
- Review article: safety and tolerability of direct-acting anti-viral agents in the era of hepatitis C therapy.Alimentary Pharmacology Therapeutics. 2016; 43: 674-696
- Hepatic decompensation during direct-acting antiviral therapy of chronic hepatitis C.Journal of Hepatology. 2016; 64: 763-765
Article info
Publication history
Published online: April 08, 2017
Accepted:
March 30,
2017
Received:
November 26,
2016
Identification
Copyright
© 2017 Editrice Gastroenterologica Italiana S.r.l. Published by Elsevier Ltd. All rights reserved.
