Patient-reported outcomes in chronic hepatitis C patients with cirrhosis treated with sofosbuvir-containing regimens


  • Potential conflict of interest: Nothing to report.


Whether the presence of cirrhosis influences patient-reported outcomes (PROs), including health-related quality of life, during treatment with newly available anti-HCV (hepatitis C virus) regimens is unclear. Our aim was to assess the association of cirrhosis with PROs in patients treated with sofosbuvir (SOF)-containing regimens. Four PRO questionnaires (Short Form-36 [SF-36], Functional Assessment of Chronic Illness Therapy-Fatigue [FACIT-F], Chronic Liver Disease Questionnaire-HCV (CLDQ-HCV), and the Work Productivity and Activity Impairment Questionnaire: Specific Health Problem [WPAI-SHP]) were administered to subjects receiving SOF and ribavirin (RBV; FUSION trial, N = 201, 34% cirrhosis; VALENCE trial: N = 333, 21% cirrhosis) and SOF, RBV, and pegylated interferon (Peg-IFN; NEUTRINO trial: N = 327, 17% cirrhosis). HCV patients with cirrhosis showed significant impairment of PROs before initiation of treatment. During treatment, patients with cirrhosis treated with the IFN-free regimen experienced moderate decline in their PRO scores (0.6%-5.2% on a normalized scale of the summary scores; all P > 0.02). In contrast, patients with cirrhosis treated with IFN-containing regimen showed decline in PRO scores that ranged from 3.4% to 16.0% (all P < 0.005). Nevertheless, by follow-up week 12, no PRO decrement from baseline was observed in patients with cirrhosis regardless of the treatment regimen. Furthermore, in patients with cirrhosis with HCV who achieved sustained virological response at 12 weeks (SVR-12), some improvement in PROs from baseline was observed. During treatment, changes in PRO scores were similar between patients with and without cirrhosis for both treatment regimens (all P > 0.05). Independent predictors of lower PROs in patients with cirrhosis included baseline depression, anxiety, fatigue, high HCV viral load, female gender, and receiving IFN-containing treatment. Conclusions: Treatment with SOF+RBV with or without Peg-IFN is tolerated by HCV patients with and without cirrhosis in terms of their PRO scores. After achieving SVR-12 with the IFN-free regimen, patients with cirrhosis showed improvement in some aspects of their PROs. (Hepatology 2014;59:2161–2169)


alanine aminotransferase


aspartate aminotransferase to platelet ratio index


body mass index




chronic hepatitis C


Chronic Liver Disease Questionnaire-HCV


Functional Assessment of Chronic Illness Therapy-Fatigue


hepatitis C virus


health-related quality of life




international normalized ratio


pegylated IFN


patient-reported outcomes




Short Form-36




sustained virological response at 12 weeks




upper limit of normal


Work Productivity and Activity Impairment Questionnaire: Specific Health Problem

Chronic hepatitis C (CHC) is the most common cause of cirrhosis and its complications (hepatocellular carcinoma) in the United States and the Western world.[1-9] CHC has been associated with increased mortality and resource utilization as well as impairment of patients' well-being.[4] Although hepatitis C virus (HCV), by itself, can negatively affect patient-reported outcomes (PROs), presence of cirrhosis, especially in its advanced form, can further deteriorate this impairment.[5-8]

Until recently, treatment of CHC was based on interferon (IFN) with substantial side effects, which could negatively affect patients' well-being.[10, 11] The recent approval of new anti-HCV regimens, including IFN-free regimens, not only has led to significant improvement in efficacy, but also has provided patients with a more tolerable regimen with shorter treatment duration. Furthermore, these regimens have been shown to be cost-effective and improve patients' health-related quality of life (HRQL) and other PROs.[12-14] Given further impairment of HRQL in patients with cirrhosis, it becomes critical to assess the effect of these regimens on PROs of CHC patients with cirrhosis who are treated with these new regimens.

Therefore, the aim of this study was to assess the effect of cirrhosis on PROs as well as the changes in PRO scores during treatment with sofosbuvir (SOF) with ribavirin (RBV) with or without IFN.

Patients and Methods

The Study Cohort

In this study, we used the PRO data for CHC patients with cirrhosis collected in three recently conducted phase III trials of SOF in combination with RBV with or without pegylated IFN (Peg-IFN): the FUSION, the NEUTRINO, and the VALENCE trials. The study design and the results of those trials have been reported on elsewhere.[15-17] Briefly, in VALENCE, treatment-naïve or experienced patients with genotype 2 and 3 HCV infection were receiving SOF 400 mg once-daily plus weight-based RBV twice-daily (1,000 or 1,200 mg/day) for 12 or 24 weeks depending on HCV genotype. In FUSION, treatment-experienced patients with HCV genotype 2 and 3 were one-to-one randomized to receive 12 or 16 weeks of the same SOF+RBV therapy. In NEUTRINO, treatment-naïve patients with HCV genotype 1, 4, 5, or 6 were receiving 12 weeks of the same SOF+RBV therapy plus Peg-IFN alpha-2a once-weekly at a dose of 180 ug. All patients were further followed up 4 weeks after completion of active treatment, and those with undetectable RNA at week 4 posttreatment were also followed up at week 12. The studies were conducted in the United States, Canada, Australia, New Zealand (FUSION and NEUTRINO), and 10 European countries (VALENCE).[15-17]

For the purpose of this study, using medical history collected at screening for all participants enrolled in the clinical trials, we identified patients with pretreatment depression, fatigue, anxiety, and insomnia. Also, in this study, treatment-related anemia was defined as a decrement in hemoglobin of at least 2 g/dL from baseline while receiving active treatment.

In this study, all patients were screened for the presence or absence of cirrhosis. The presence of cirrhosis was defined as any one of the following: liver biopsy showing cirrhosis; FibroScan (in countries where locally approved) showing cirrhosis or results >12.5 kPa9; a FibroTest score of >0.75 and an aspartate aminotransferase to platelet ratio index (APRI) of >2 during screening. The absence of cirrhosis was defined as any one of the following: liver biopsy within 2 years of screening showing absence of cirrhosis; FibroScan with a result of ≤12.5 kPa within ≤6 months of baseline; a FibroTest score of ≤0.48; and an APRI of ≤1 during screening. In the absence of a definitive diagnosis of the presence or absence of cirrhosis by the above criteria, a liver biopsy was required. Liver biopsy results superseded the results obtained by FibroScan or FibroTest.

Patients with a current or previous history of clinical hepatic decompensation (e.g., ascites, jaundice, encephalopathy, or variceal hemorrhage) were considered ineligible for these studies.

Measurement of PROs

Four PRO questionnaires—the Short Form-36 (SF-36), Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F), Chronic Liver Disease Questionnaire-HCV (CLDQ-HCV), and the Work Productivity and Activity Impairment Questionnaire: Specific Health Problem (WPAI-SHP)—were administered to the study participants in their native language at baseline, week 4 (FUSION and VALENCE only), the last day of treatment (week 12, 16, or 24 depending on the study), and weeks 4 and 12 posttreatment. The questionnaires were self-administered by the trials' participants in a clinic room before initiation of any other study-related activities. Both patients and site staff were blinded to patients' HCV RNA level measured at the time of each visit.

The SF-36 version 2 is a widely used instrument for HRQL evaluation.[18] It assesses eight HRQL scales (ranging from 0 to 100, with higher values corresponding to a better health status): physical functioning (PF); role physical (RP); bodily pain (BP); general health (GH); vitality (VT); social functioning (SF); role emotional (RE); and mental health (MH). The two summary scores summarize the physical and mental health components of SF-36: the Physical Component Summary score (PCS) and Mental Component Summary score (MCS). In this study, the SF-36 scales and summary scores were calculated using QualityMetric Health Outcomes Scoring Software (4.5; QualityMetric, Lincoln, RI) and the 2009 U.S. population norms.[18]

The CLDQ-HCV is another widely used and validated HRQL instrument developed specifically for assessment of HRQL in HCV patients.[19] It includes four HRQL scales: activity and energy (AE); emotional (EM); worry (WO); and systemic (SY). These scales are averaged to the total CLDQ-HCV score that ranges from 1 to 7, with higher values representing better HRQL.

The FACIT-F is a widely used and extensively validated 40-item PRO questionnaire that assesses fatigue and its effect upon daily activities.[20] The scoring scheme includes physical (PWB; range, 0-28), emotional (EWB; range, 0-24), social (SWB; range, 0-28), and functional (FWB; range, 0-28) well-being domains as well as the fatigue subscale domain (FS; range, 0-52); these five scales together add up to the total FACIT-F score that ranges from 0 to 160, with higher values representing better well-being.

Finally, the WPAI-SHP is another validated PRO instrument where participants are asked to evaluate impairment in their daily activities and work productivity associated with a specific health problem.[21] In this study, participants were specifically asked about the effect of their HCV infection. In the WPAI-SHP, the work impairment domain is a sum of impairment in work productivity resulting from absenteeism and impairment stemming from decreased productivity while working (presenteeism); this domain is assessed only in those who report being employed at the time of completing the questionnaire. The activity impairment domain represents impairment in daily activities other than work and is assessed in all participants regardless of their employment status. In this instrument, unlike three others, higher impairment scores indicate poorer health status: The minimum possible value of 0 represents no impairment in work productivity or daily activities, whereas the value of 1 represents complete inability to work or perform those activities.

Statistical Analysis

Clinicodemographic parameters and all HRQL and PRO scores were compared between patients with and without cirrhosis at all time points using Pearson's chi-square test for independence or Wilcoxon's nonparametric test. In this study, we also calculated decrements in PROs from patients' own baseline and used Wilcoxon's sign-rank test for matched-pairs comparison to identify the decrements that were significant. P values of 0.05 or less were considered statistically significant.

In patients with cirrhosis, independent predictors of PROs at different time points were assessed using multiple linear regression. In the IFN-free study arm, duration of treatment with SOF+RBV was used as one of the potential predictors. Other potential PRO predictors were age, gender, ethnicity, body mass index (BMI), baseline hemoglobin (at day 1) or treatment-related anemia (after the start of treatment), pretreatment history of psychiatric disorders, including depression, anxiety, fatigue, and insomnia, baseline HCV viral load and alanine aminotransferase (ALT), and achieving sustained virological response at 12 weeks (SVR-12; posttreatment only).

In this study, at all time points after the start of active treatment, we merged together all four IFN-free treatment regimens studied in the FUSION and VALENCE; this merged study cohort is further referred to as the IFN-free study arm. The IFN-containing regimen (the NEUTRINO study) was analyzed separately.

These studies had been separately approved by each site's institutional review board.


Baseline PROs in Patients With and Without Cirrhosis

Of 852 enrolled patients with CHC, 192 had compensated cirrhosis before initiation of treatment. The demographic characteristics of patients with and without cirrhosis are shown in Table 1. Patients with cirrhosis were significantly older, more likely male, and treatment experienced (all P < 0.05). Furthermore, patients with cirrhosis had higher baseline BMI and higher proportion of those with high ALT (above 1.5 gender-defined upper limits of normal [ULNs]), lower baseline serum albumin and platelets, and higher international normalized ratio (INR; Table 1). Other clinicodemographic parameters of patients with and without cirrhosis were similar, including the rates of baseline history of depression, anxiety, insomnia, and fatigue (all P > 0.05).

Table 1. Baseline Clinicodemographic Profile of the Study Participants
 CirrhosisNo CirrhosisP Value
  1. Abbreviation: NS, not significant.

Age, years54.9 ± 7.150.7 ± 10.6<0.0001
Male (%)134 (69.8)408 (61.8)0.04326
Caucasian (%)167 (87.0)566 (85.8)NS
Treatment naive (%)69 (35.9)394 (59.7)<0.0001
HCV load >106 (%)139 (72.4)496 (75.2)NS
ALT >1.5× ULN (%)151 (78.6)353 (53.5)<0.0001
Hemoglobin, g/dL14.87 ± 1.2714.79 ± 1.34NS
Albumin, g/dL3.85 ± 0.384.05 ± 0.30<0.0001
INR1.06 ± 0.101.00 ± 0.23<0.0001
Platelets, ×106/mL147.8 ± 52.6223.3 ± 64.1<0.0001
BMI28.4 ± 4. 827.4 ± 5.40.0013
Anxiety (%)24 (12.5)81 (12.3)NS
Depression (%)44 (22.9)130 (19.7)NS
Symptomatic fatigue (%)22 (11.5)59 (8.9)NS
Insomnia (%)33 (17.2)109 (16.5)NS

Baseline PROs of patients with and without cirrhosis are shown in Table 2. Despite similar rates of baseline depression and other potentially PRO-related disorders, nearly all PROs, including individual and summary scales, were lower in patients with cirrhosis; the magnitude of difference ranged from 3.2% to 9.3% on a normalized 0%-100% scale for all significantly different PROs. The only baseline PRO that was not different between patients with and without cirrhosis was work productivity measured by both presenteeism and absenteeism (P > 0.05; Table 2).

Table 2. Baseline Health-Related Quality of Life and Patient-Reported Outcomes in Patients With and Without Cirrhosis
 CirrhosisNo CirrhosisP Value
Physical  functioning79.21 ± 22.8183.07 ± 22.080.0051
Role physical71.58 ± 29.6879.23 ± 25.560.0014
Bodily pain70.39 ± 26.0574.52 ± 25.440.0478
General health58.09 ± 22.0565.16 ± 21.510.0001
Vitality57.23 ± 23.2762.22 ± 22.350.0115
Social functioning77.03 ± 24.5781.83 ± 23.700.0095
Role emotional77.62 ± 27.9283.30 ± 22.790.0415
Mental health69.13 ± 19.8373.16 ± 18.590.0122
Physical summary scale49.14 ± 9.0651.14 ± 8.750.0031
Mental summary scale47.61 ± 10.8749.85 ± 10.020.0166
Physical well-being22.70 ± 4.9623.60 ± 4.810.0153
Social well-being20.94 ± 5.6221.49 ± 6.150.0567
Emotional well-being17.01 ± 4.3118.31 ± 3.990.0002
Functional well-being19.017 ± 6.3620.55 ± 5.680.0061
Fatigue36.93 ± 12.5840.22 ± 11.050.0024
Total FACIT-F116.69 ± 26.92124.18 ± 25.640.0006
Activity/energy5.00 ± 1.465.43 ± 1.280.0007
Emotional5.14 ± 1.195.44 ± 1.150.0018
Worry4.97 ± 1.415.51 ± 1.18<0.0001
Systemic4.80 ± 1.325.22 ± 1.270.0002
Total CLDQ-HCV4.98 ± 1.175.40 ± 1.05<0.0001
Work productivity impairment0.150 ± 0.2450.099 ± 0.1960.0611
Absenteeism0.046 ± 0.1560.026 ± 0.1150.1185
Presenteeism0.105 ± 0.1740.075 ± 0.1490.0534
Activity impairment0.241 ± 0.2840.148 ± 0.2290.0001

Treatment of HCV Patients With Cirrhosis With IFN-Free Regimens

A total of 528 CHC patients were enrolled in the IFN-free treatment arm; 26.1% had cirrhosis. Of all the patients, 183 were treated with SOF and RBV for 12 weeks (100 in FUSION and 83 in VALENCE), 95 for 16 weeks (FUSION only), and 250 for 24 weeks (VALENCE only).

At week 4 of active treatment, a decline in PROs was observed in both patients with and without cirrhosis (Fig. 1), although, similarly to baseline, many PRO scores remained lower in patients with cirrhosis (47.7 vs. 49.7, P = 0.0424 for PCS; 34.8 vs. 37.3, P = 0.0186 for FS; 111.0 vs. 116.5, P = 0.0348 for FACIT-F; 4.79 vs. 5.23, P = 0.0008 for CLDQ-HCV). By the end of treatment, a more substantial decline was observed in most of the PROs, regardless of patients' cirrhosis status (Fig. 1). Nevertheless, the actual values of FS and FACIT-F were no longer different between patients with and without cirrhosis (P > 0.05; Fig. 1), whereas PCS and CLDQ-HCV remained lower in patients with cirrhosis (47.5 vs. 49.8, P = 0.0181 and 4.87 vs. 5.18, P = 0.0192, respectively).

Figure 1.

PROs in the IFN-free arm: (A) SF-36, (B) FACIT-F, (C) CLDQ-HCV, and (D) WPAI-SHP. Dark blue, dark green lines: cirrhosis; light blue, light green: no cirrhosis. P values represent the difference between patients with and without cirrhosis at baseline, end of treatment (EoT), and posttreatment week 12 (F/U w12).

However, despite these differences in the absolute PRO values, no PRO metric, neither an individual nor a summary, experienced a more substantial decrement in patients with cirrhosis during the course of treatment with SOF+RBV (all P > 0.05, when compared to patients without cirrhosis).

At follow-up, in patients with cirrhosis, all PROs returned to their baseline levels or moderately improved as early as posttreatment week 4 (Fig. 1). At week 12 follow-up, all PROs returned to their baseline levels, regardless of cirrhosis status (Fig. 1). Furthermore, regardless of the presence of cirrhosis, moderate improvements from baseline were observed for some of the studied PROs (FS and total FACIT-F, as well as CLDQ-HCV). The PCS also demonstrated a moderate improvement in patients without cirrhosis, but not in patients with cirrhosis, but it is unclear whether this could have occurred as a result of the difference in sample size, because direct comparison of the average improvements in PCS between those with and without cirrhosis did not reveal statistical significance (0.89 ± 6.68 vs. 1.31 ± 6.00; P > 0.05). Finally, of patients with cirrhosis who achieved SVR-12 (N = 80), a statistically significant improvement from baseline was observed in vitality of SF-36 (by 5.98 ± 19.57; P = 0.0189), emotional well-being (2.12 ± 4.65; P = 0.0004), fatigue (3.15 ± 9.55; P = 0.0129), and total of FACIT-F (5.34 ± 18.19; P = 0.0222), as well as most of the scales of CLDQ-HCV (by 0.32-0.63; P = 0.0391-0.0001; Supporting Table 1).

Treatment of HCV Patients With Cirrhosis With an IFN-Containing Regimen

A total of 324 CHC patients were enrolled in the IFN-containing treatment arm. The prevalence of cirrhosis was 16.7% (N = 54). All patients were treatment naïve.

In this treatment arm, the actual values of PROs were not different between those with and without cirrhosis at all time points (Fig. 2). This includes the baseline time point where the patients had similar PROs regardless of their cirrhosis status (all P > 0.05).

Figure 2.

PROs in the IFN-containing arm: (A) SF-36, (B) FACIT-F, (C) CLDQ-HCV, and (D) WPAI:SHP. Dark blue, dark green lines: cirrhosis; light blue, light green: no cirrhosis. P values represent the difference between patients with and without cirrhosis at baseline, end of treatment (EoT), and posttreatment weeks 4 and 12 (F/U w4, F/U w12).

By the end of treatment, substantial decrements in all PROs were experienced by treated patients regardless of their cirrhosis status (Fig. 2; all P < 0.05). In fact, the decrements from baseline in PF (16.6 vs. 8.5; P = 0.0176), BP (14.9 vs. 5.7; P = 0.0190), and PCS (6.4 vs. 3.4; P = 0.0332) were larger in patients with no cirrhosis, as compared to those with cirrhosis. Furthermore, similarly to IFN-free treatment, no PRO metric experienced more substantial decrement during treatment in patients with cirrhosis, when directly compared to patients without cirrhosis (all P > 0.05, except for those presented above).

Four weeks later, some of the PROs returned to baseline levels in both patients with and without cirrhosis (FS, total FACIT-F, and CLDQ-HCV), whereas others remained moderately impaired. Furthermore, at week 12 follow-up, moderate improvements with reference to patients' own baseline were observed in both cirrhosis and noncirrhosis groups (Fig. 2). Among those with cirrhosis who achieved SVR-12 (N = 43), statistically significant improvements from baseline were observed in EWB (by 2.24 ± 4.31; P = 0.0023) and FS (3.41 ± 8.76; P = 0.137) of FACIT-F and all scales of CLDQ-HCV, except for systemic well-being (by 0.37 to 0.54; all P < 0.05; Supporting Table 1). This pattern of improvement was similar to that reported for the IFN-free treatment.

Independent Predictors of PROs in Patients With Cirrhosis

In the initial multivariate analysis, the two treatment regimens and all patients, regardless of cirrhosis status, were merged together (Supporting Table 2). The presence of cirrhosis, type of treatment, and baseline clinicodemographic parameters were among the variables considered. The multivariate analysis showed that, in addition to baseline clinicodemographic variables (female gender, baseline depression, anxiety, insomnia, fatigue, and having a history of unsuccessful treatment), having cirrhosis was also associated with more impairment in PROs documented by FS (beta = −2.22 ± 0.93; P = 0.0177), FACIT-F (−4.84 ± 2.15; P = 0.0250), and CLDQ-HCV (−0.354 ± 0.090; P = 0.0001) scores as well as the activity impairment domain of WPAI (0.077 ± 0.021; P = 0.0002). Furthermore, during treatment, receiving an IFN-containing regimen was another independent predictor of PRO impairment (normalized betas: −3.4% to −17.3%; P < 0.02; Supporting Table 2), whereas having cirrhosis was no longer associated with any of the PRO impairment during treatment (all P > 0.05). Finally, 12 weeks after the end of treatment, having cirrhosis returned as an independent predictor of the following PROs: PCS (beta = −2.69 ± 0.81; P = 0.0009), CLDQ-HCV (−0.339 ± 0.098; P = 0.0006), work productivity (0.092 ± 0.029; P = 0.0017), and activity impairment (0.073 ± 0.021; P = 0.0006; Supporting Table 2).

In a separate multivariate analysis that focused on patients with cirrhosis who were treated with an IFN-free regimen, baseline depression, anxiety, fatigue, insomnia, and female gender were consistent predictors of lower PROs before, during, and after treatment with SOF+RBV (Table 3A).

Table 3. Independent Predictors of PROs in Patients With Cirrhosis
 A. IFN-Free ArmB. IFN-Containing Arm
 PredictorBeta ± SEabP ValuePredictorBeta ± SEP Value
  1. a

    Beta indicates the magnitude of change in PRO if a predictor changes by the magnitude of 1 (for a binary predictor, is present compared to its absence) given that all other predictors remain fixed.

  2. b

    Associations with better PROs are defined by positive beta values for SF-36, FACIT-F, and CLDQ-HCV scores; negative beta values for work productivity and activity impairment scores.

  3. Abbreviation: SE, standard error.

Physical summary score (SF-36)Depression−7.11 ± 1.830.0002HCV RNA > 106−6.34 ± 2.750.0256
Mental summary score (SF-36)Treatment duration, per week−0.73 ± 0.16<0.0001Depression−7.00 ± 2.680.0119
Male gender4.92 ± 1.940.0125   
Anxiety−11.42 ± 3.490.0014   
Depression−6.86 ± 2.220.0025   
Fatigue (FACIT-F)Anxiety−9.71 ± 4.400.0294HCV RNA > 106−7.54 ± 3.270.0254
Depression−9.66 ± 2.510.0002   
Fatigue−12.2 ± 3.400.0005   
Total FACIT-FAnxiety−25.35 ± 9.810.0110HCV RNA > 106−15.33 ± 6.940.0319
Depression−20.01 ± 5.610.0005   
Fatigue−17.83 ± 7.590.0205   
CLDQ-HCVTreatment duration, per week−0.048 ± 0.0190.0143HCV RNA > 106−0.764 ± 0.2970.0131
Male gender0.570 ± 0.2230.0119   
Anxiety−1.502 ± 0.4360.0008   
Depression−0.544 ± 0.2570.0365   
Work productivity impairmentALT >1.5 × ULN−0.219 ± 0.0800.0086Insomnia0.357 ± 0.1090.0032
Treatment naïve0.259 ± 0.0930.0075   
Activity impairmentMale gender−0.153 ± 0.0580.0100HCV RNA > 1060.161 ± 0.0720.0304
Anxiety0.233 ± 0.1160.0461   
Depression0.128 ± 0.0630.0456   
Fatigue0.315 ± 0.0820.0002   
Hemoglobin, per g/dL0.056 ± 0.0230.0163   
End of treatment      
Physical summary score (SF-36)Male gender6.62 ± 1.680.0001Age, per year−0.41 ± 0.170.0186
Anxiety−8.14 ± 2.930.0064   
Fatigue−6.82 ± 2.690.0126   
Mental summary score (SF-36)Treatment duration, per week−0.57 ± 0.190.0034Depression−8.34 ± 3.160.0113
Depression−9.54 ± 2.490.0002   
Fatigue (FACIT-F)Male gender7.38 ± 2.280.0016No predictors  
Anxiety−9.20 ± 4.090.0262   
Depression−10.49 ± 2.590.0001   
Total FACIT-FMale gender15.46 ± 5.680.0074Caucasian−22.61 ± 9.670.0235
Depression−23.12 ± 6.250.0003   
CLDQ-HCVMale gender0.940 ± 0.218<0.0001No predictors  
Insomnia−0.710 ± 0.2720.0102   
Work productivity impairmentMale gender−0.162 ± 0.0790.0455Age, per year0.010 ± 0.0050.0435
Anxiety0.394 ± 0.1880.0406   
Activity impairmentMale gender−0.215 ± 0.0570.0003Age, per year0.015 ± 0.0060.0145
Insomnia0.200 ± 0.0720.0063   
Week 12 posttreatment      
Physical summary score (SF-36)Age, per year−0.24 ± 0.110.0315No predictors  
Fatigue−8.92 ± 3.320.0089   
Mental summary score (SF-36)Treatment duration, per week−0.52 ± 0.230.0293Anxiety−10.08 ± 3.480.0060
Depression−6.77 ± 3.370.0477   
Fatigue (FACIT-F)Depression−11.66 ± 2.56<0.0001Anxiety−6.46 ± 3.060.0408
Total FACIT-FDepression−17.37 ± 6.680.0110Anxiety−22.19 ± 8.300.0108
CLDQ-HCVMale gender0.599 ± 0.2260.0097Anxiety−0.96 ± 0.370.0142
Work productivity impairmentDepression0.477 ± 0.1440.0023No predictors  
Fatigue0.212 ± 0.0960.0340   
Activity impairmentMale gender−0.144 ± 0.0550.0102No predictors  

Finally, in multivariate analysis focusing on patients with cirrhosis treated with an IFN-based regimen (Table 3B), a major predictor of lower baseline PROs was high baseline viral load. During treatment, older age was associated with lower PCS, work productivity, and activity (P < 0.05), whereas after the end of treatment, the only identified predictor of lower PROs was having a history of anxiety (Table 3B).


In this study, we address two important issues related to PROs in patients with HCV-related cirrhosis. First, we attempt to ascertain the effect of cirrhosis on PROs before initiation of any anti-HCV treatment-related activities. Our data show that baseline HRQL, measured by both the generic SF-36 and the disease-specific CLDQ-HCV, documents significant impairment of HRQL in CHC patients with cirrhosis. The most profound decrements in PROs were related to activity, energy, vitality, and fatigue issues represented by the role physical and vitality domains of SF-36 and the respective domains of CLDQ-HCV and FACIT-F. Given these data, it is not unexpected that even compensated cirrhosis negatively influences issues related to work productivity measured by both presenteeism while working and performing activities other than work (both measured by the WPAI instrument). In fact, CHC patients with cirrhosis had lower PRO scores even after adjustment for baseline demographic data, such as age and gender.

The second important issue addressed by our study was related to the effect of SOF-containing regimens on PROs of CHC patients with cirrhosis during treatment and after achieving SVR. Our data show that IFN-free regimens have a modest negative effect on PROs, and this effect of treatment is not higher in patients with cirrhosis, when compared to patients without cirrhosis. Furthermore, these impairments in PROs are resolved shortly after discontinuation of the IFN-free regimen. Patients with cirrhosis who achieve SVR with IFN-free regimens show significant improvement in vitality, fatigue, activity/energy, worry, and other disease-specific aspects of HRQL. This is important because patients with cirrhosis at baseline suffer from substantial impairment of patient-reported outcomes. Furthermore, viral eradication may play an important role in improving these outcomes that are critical from patients' perspective.

Looking at the same variables for CHC patients with cirrhosis treated with SOF in an IFN-containing regimen, the decline in PRO scores was expectedly more profound, as compared to the IFN-free arm. Given the side-effect profile of IFN, these impairments are expected and are also similar to the previously reported values.[13, 14, 22, 23] Nevertheless, the relatively short duration of this treatment (12 weeks) and complete posttreatment recovery of all the PRO scores in patients with cirrhosis suggest that the IFN-containing SOF-based regimen can be well tolerated by patients with HCV with cirrhosis. Although some improvement was noted in patients with cirrhosis who achieved SVR with the IFN-containing regimen, these changes were not as consistent as those observed in patients with cirrhosis who achieved SVR-12 with the IFN-free regimens. This supports the notion that IFN-related side effects may take longer than 12 weeks to fully resolve. In fact, the role emotional component of SF-36 continued to show some deficit in patients with cirrhosis with SVR as late as 12 weeks after the end of treatment. This is consistent with the long-term side-effect profile of IFN exemplified by the reports of cases of depression months after discontinuation of IFN.[23, 24]

Our study also assessed predictors of PROs in CHC patients with cirrhosis who were treated with IFN-free or IFN-containing regimens. As previously reported, depression, fatigue, and anxiety are the most consistent independent predictors of PROs at any time point.[23-26] In a pooled multivariate analysis, having cirrhosis was also associated with lower PROs while being off-treatment (i.e., at baseline or during posttreatment follow-up). On the other hand, during treatment, having cirrhosis was no longer associated with additional PRO impairment. This gives further support to the premise that well-compensated CHC patients with cirrhosis can tolerate SOF-containing regimens in a similar fashion to CHC patients without cirrhosis.

Our data also suggest that patients with cirrhosis with low albumin and platelets also can tolerate SOF-containing treatment reasonably well. In fact, 17 CHC patients with cirrhosis had the reported unfavorable safety profiles for treatment with triple combinations containing the first generation direct antiviral agents (DAAs) (albumin below 3.5 g/dL and platelets below 100,000/mL).[27] Nevertheless, in our study, for these patients, the rate of treatment-related anemia was significantly lower (47.1% vs. 82.9% in other patients with cirrhosis; P = 0.0005), likely because of lower baseline hemoglobin (14.0 ± 1.3 vs. 15.0 ± 1.2; P = 0.0040), whereas other non-PRO parameters and baseline PROs were similar, as was the SVR rate (all P > 0.05). Because only 2 of those patients were treated with the IFN-containing regimen, we studied on-treatment PROs only in those receiving the IFN-free treatment. In those patients, of their PROs during treatment, only PF and PCS of SF-36 declined more substantially, compared to other patients with cirrhosis: by −17.1 versus −3.4 (P = 0.0229) and by −4.38 versus −0.87 (P = 0.0269). Both of these aspects of PROs returned to their baseline levels as early as week 4 posttreatment and were no longer different from the rest of the patients with cirrhosis. Thus, patients with compensated cirrhosis who were advanced enough to make them ineligible for triple therapy with IFN and first generation DAAs, without compromising their PROs.

To the study limitations, the data used for this analysis came from the multicenter clinical trials, so these results may not necessarily be generalizable to all facilities or the general population of patients with cirrhosis with CHC. In fact, patients with decompensated cirrhosis were excluded from these trials, whereas the effect of anti-HCV treatment may be substantially different for these populations; further studies are needed to assess this effect in those patients. We also could not contrast post-SVR PRO improvements to those who did not achieve SVR because of the lack of the necessary follow-up data. Nevertheless, this is the largest PRO study in patients with cirrhosis with CHC treated with IFN-free and IFN-containing regimens. The careful assessment of PROs with multiple validated instruments and longitudinal assessments provides substantial information about outcomes that are important to patients with CHC.

In conclusion, patients with CHC and cirrhosis who participated in the recent phase III clinical trials of SOF-based treatment experienced some decrement in their PROs before initiation of treatment. However, IFN-free regimens were associated with minimal PRO decrements. Although PROs were more substantially affected by the IFN-containing SOF-based treatment, these impairments were similar between patients with and without cirrhosis, whereas the short duration of treatment (12 weeks) does provide advantage to these patients. Finally, patients with cirrhosis who achieved SVR-12 after treatment, especially with the IFN-free SOF-based regimens, enjoyed significant improvement of some of their PRO scores. It is important that future studies of decompensated patients with cirrhosis treated with IFN-free regimens are to be assessed not only for efficacy and safety, but also for the effect of these regimens on PROs.