Association between Liver Stiffness and Liver-Related Events in HCV-Infected Patients after Successful Treatment with Direct-Acting Antivirals

Background and Objectives: Direct-acting antivirals (DAAs) are highly effective for the treatment of chronic hepatitis C virus (HCV) infection, but the risk of liver-related events and hepatocellular carcinoma (HCC) remains after successful therapy. We aimed to evaluate post-treatment changes in liver stiffness (LS) and identify a cut-off LS value for predicting such events in chronic HCV-infected patients receiving DAA. Materials and Methods: A total of 185 patients who had achieved sustained virologic response (SVR) after DAA therapy were included. Baseline characteristics and laboratory results were retrospectively abstracted. LS was measured by transient elastography at baseline, 12, 24, 48, and 96 weeks after SVR. FIB-4 index was assessed at baseline and 48 weeks after SVR. Development of liver-related events (hepatocellular carcinoma (HCC), portal-hypertension-related decompensation, listing for transplantation, and mortality) after SVR were identified. The association between liver fibrosis and the occurrence of liver-related events was analyzed using Cox regression analysis. Results: Significant differences in LS values were observed between baseline and 24, 48, 72, and 96 weeks after SVR. FIB-4 index at 48 weeks after SVR was significantly lower than the FIB-4 index at baseline. During the 41.6-month follow-up time, the incidence rates of all liver-related events and HCC were 2.36 and 1.17 per 100 person-years, respectively. Age, LS ≥8 kPa, and FIB-4 ≥1.35 at 48 weeks post-SVR were significantly associated with the occurrence of any liver-related events. By multivariate analysis, LS ≥8 kPa at 48 weeks post-SVR remained significantly associated with any liver-related events, with an adjusted hazard ratio (95%CI) of 5.04 (1.01–25.26), p = 0.049. Conclusions: Despite a significant reduction in LS after SVR, patients with LS ≥8 kPa at 48 weeks after SVR should be regularly monitored for liver-related complications, particularly HCC development.


Introduction
Chronic hepatitis C virus (HCV) infection is one of the common causes of chronic liver disease [1]. In 2019, an estimated 58 million people were chronically infected with HCV, resulting in 290,000 deaths worldwide. If left untreated, chronic HCV infection causes persistent hepatic inflammation, leading to progressive hepatic fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) [2,3].
Direct-acting antivirals (DAAs) are currently a standard treatment for chronic HCV infection. DAAs specifically inhibit the synthesis of HCV-encoded proteins necessary for viral replication. DAAs have higher rates of sustained virologic response (SVR) and more favorable safety profiles than traditional interferon-based regimens [4,5]. In a study of 102 HCV-infected patients treated with DAA, DAA therapy was significantly associated with liver fibrosis regression [6]. Another study consistently reported that patients who achieved SVR with DAA treatment had significantly reduced liver fibrosis measured by transient elastography (TE), i.e., from a median liver stiffness (LS) of 8.3 kilopascals (kPa) at baseline to 5.4 kPa at 48 weeks after the end of treatment [7]. Achieving SVR was found to be significantly associated with lower risks of all-cause mortality over an 8.4-year follow-up period [8]. Patients with SVR had a markedly lower 10-year cumulative HCC incidence rate than those without SVR, 5.1% vs. 21.8%, respectively [8].
Despite the high efficacy of DAAs, liver fibrosis regression after SVR is not always guaranteed, and the risk of liver-related events, particularly hepatocellular carcinoma (HCC), is not completely eliminated, with an increasing number of reports of HCC development following successful therapy. At 96 weeks after successful DAA treatment, 17% of patients had progressive hepatic fibrosis [9]. In another observational study, the progression of liver fibrosis occurred in 12.5% of patients achieving SVR post-DAA therapy [10]. Notably, the incidence rate of HCC was the highest among patients with fibrosis progression (6.17/100 patient-years) compared to patients with stable fibrosis (1.09/100 patient-years) and those with fibrosis regression (0.75/100 patient-years) [10]. These findings supported an observation in a previous cohort study showing that patients with progressive liver fibrosis after SVR had a significantly higher rate of HCC development than those with fibrosis regression or stability (33% vs. 4% at 5 years, p < 0.001) [11]. After achieving SVR, the annual incidence of HCC was reported to be 1.5 per 100 patient-years in patients with HCV-associated compensated advanced chronic liver disease, and the incidence increased to 1.97 per 100 patient-years in patients with HCV-related cirrhosis [12,13]. These findings highlight the importance of long-term evaluation and monitoring of liver fibrosis in determining the risk of HCC after SVR.
A liver biopsy is the gold standard for diagnosing and staging liver fibrosis. However, it is invasive, with potentially fatal complications and sampling error [14]. As a result, several non-invasive tests, such as the TE and Fibrosis-4 (FIB-4) index, which indirectly reflects the degree of liver fibrosis, have been used to assess liver fibrosis [15,16]. According to international guidelines, following SVR, cirrhotic patients should have an ultrasound examination every 6 months for HCC surveillance, but monitoring for liver-related events or HCC occurrence is currently not recommended for non-cirrhotic patients [17,18]. Although TE and FIB-4 index is currently used to evaluate liver fibrosis, it has not yet been recommended for the post-treatment monitoring of chronic HCV patients due to the lack of consensus on a cut-off value for liver fibrosis that predicts liver-related events in patients who have achieved SVR [19]. This study aimed to evaluate the post-treatment changes in liver stiffness (LS) up to 96 weeks after SVR and to determine the cut-off of LS associated with liver-related events, particularly HCC development, for chronic HCV-infected patients after SVR with DAA therapy.

Patients
All patients with chronic HCV infection who received DAA therapy at the King Chulalongkorn Memorial Hospital, Bangkok, Thailand, between January 2016 and December 2020 were identified (n = 864). The inclusion criteria were as follows: (i) aged 18-70 years, (ii) having completed the DAA regimen and achieved SVR, and (iii) having LS measurement by TE prior to starting DAA and at least one of the following time points: 24, 48, 72, and 96 weeks after achieving SVR. Patients were excluded if they had concomitant etiologies of chronic liver disease, e.g., chronic hepatitis B or human immunodeficiency virus infection, or if they had been diagnosed with HCC or terminal stage of any disease such as end-stage renal disease.

HCV Treatment
The regimen and duration of DAA therapy for each patient were determined by the American Association for the Study of Liver Diseases (AASLD) guidelines, along with the expert opinion of hepatologists at our institution. During the study period, several sofosbuvir-based DAA regimens were available and prescribed to patients, including sofosbuvir/daclatasvir, sofosbuvir/ledipasvir, sofosbuvir/velpatasvir, and sofosbuvir/interferon. Ribavirin was given to patients with cirrhosis and those who had previously failed therapy.

Data Collection
We retrospectively abstracted clinical and laboratory data prior to starting DAA treatment from electronic medical records, including age, gender, body mass index (BMI), comorbidities, presence of cirrhosis, history of HCV treatment (naïve vs. experienced treatment), liver function test, complete blood count, HCV RNA viral load, and genotype. HCV RNA viral load was measured at baseline and 12 weeks after the end of treatment by real-time PCR. The HCV genotype was determined using the linear array method. SVR was defined as undetectable HCV RNA in the blood at 12 weeks or more after completing the DAA regimen. The diagnosis of cirrhosis was made by liver histology and/or radiologic evidence of small nodular surface liver with portal hypertension (e.g., collateral vessels and splenomegaly).
LS was measured before starting DAA and after completing the regimen by TE with Fibrocan (FibroScan ® 502 Touch, Echosens, Paris, France). Liver steatosis was measured by controlled attenuation parameter (CAP) with Fibroscan. The median LS and CAP values were expressed in kPa and decibels per meter (dB/m), respectively. Details of the technical background and examination procedure have been previously described [20]. The FIB-4 index was calculated using Sterling's formula [16].
The follow-up period began at the date of SVR and ended when any liver-related events occurred. The occurrence of liver-related events was defined as death from any causes (liver and non-liver-related mortality), being listed for liver transplantation, being diagnosed with HCC, and having portal hypertension-related liver decompensation. If no event occurred, the follow-up ended in December 2020. The diagnosis of HCC was made using dynamic contrast radiologic criteria or histology when available [21]. Portal hypertension-related liver decompensation was defined as an episode of either ascites development, variceal bleeding, or overt hepatic encephalopathy. Death caused by portal hypertension-related liver decompensation or HCC was considered liver-related mortality, while death from other causes was classified as non-liver-related mortality.

Statistical Analysis
Quantitative variables were expressed as mean with standard deviation (SD) or median with range or interquartile range (IQR) as appropriate. Qualitative variables were expressed as absolute frequencies and percentages. The difference between LS and CAP at baseline and LS and CAP at 24, 48, 72, and 96 weeks after the end of DAA treatment was calculated to determine the magnitude of changes in LS and CAP after successful DDA therapy. The difference in the FIB-4 index between baseline and 48 weeks after SVR was also assessed. The incidence of liver-related events was counted and displayed as a rate per 100 person-years. In the case that a patient experienced more than one liver-related event, the first event was used for statistical analysis. A Cox regression analysis was applied to determine the association of the LS value and the FIB-4 index at various time points and the occurrence of liver-related events. Other factors associated with liver-related events were also identified. Factors significantly associated with an incident liver-related event identified in univariate analysis were included in multivariate analysis adjusted for age and gender. All statistical analyses were performed by IBM SPSS Statistics version 28.0.0 (IBM Corp, Armonk, NY, USA). A p-value of 0.05 was considered significant.

Baseline Characteristics of the Study Cohort
A total of 185 chronic HCV-infected patients met the selection criteria and were included in the analysis. Table 1 summarizes the baseline characteristics of the patient cohort. There were 115 (62.2%) males, with a mean age (± SD) of 56.1 ± 10.9 years. Cirrhosis was present in 80 (43.2%) patients prior to DAA treatment. Genotype 1 was the most common genotype (n = 100, 54%), followed by genotypes 3 and 6. Most patients (n = 130, 70.3%) were naïve to HCV treatments. Sofosbuvir/daclatasvir/ribavirin was the most commonly prescribed DDA regimen in this cohort (n = 53, 28.7%). The percentages of patients receiving each DAA regimen are shown in Table 1. Nearly all patients (n = 175, 94.6%) were prescribed a 12-week DAA regimen.  Table 2 displays the LS values at each follow-up time point. LS value significantly decreased from 12 kPa at baseline to 8.5 kPa at 24 weeks after SVR, p < 0.001. The LS value progressively declined to 5.8 kPa at 48 weeks post-SVR, p < 0.001. The LS decreased to a lesser extent at 72 weeks and 96 weeks post-SVR, but the values remained statistically significant difference from the baseline value (Table 2).

FIB-4 Index at Baseline and 48 Weeks after SVR
Similar to the changes in LS value after SVR, a significant decrease in the FIB-4 index was observed. The median FIB-4 index was 2.08 (IQR: 1.21-4.00) at baseline and 0.55 (IQR: 0.05-2.03) at 48 weeks after SVR, p < 0.001.

Liver-Related Events after SVR with DAA Therapy
During the mean follow-up time of 41.6 months after SVR, the incidence rate of any liver-related events was 2.36 per 100 person-years (Table 3). Six patients developed HCC, accounting for the 1.17 per 100 person-years HCC incidence rate (Table 4). All six patients had radiologic examinations to confirm the absence of HCC prior to HCV treatment. Two patients experienced liver decompensation as a result of portal hypertension (0.39 per 100 person-years). One had variceal bleeding and ascites on the same visit, and the other had variceal bleeding only. One patient had been listed for liver transplantation (0.19 per 100 person-years). None of the patients had died from liver-related causes, while five died from non-liver-related causes (0.95 per 100 person-years), i.e., two from COVID-19 infection, two from sepsis, and one from lung cancer.

Factors Associated with Liver-Related Events
We found that the LS values at 24 and 72 weeks after SVR were not significantly different from the value at 48 weeks after SVR (p = 0.18 and 0.86). To maximize the power of this analysis, we used their LS values at 24 or 72 weeks post-SVR for those who did not have an LS value at 48 weeks post-SVR available, i.e., LS value at 48 ± 24 weeks, yielding a total of 169 patients for this analysis.
In univariate analysis, age, LS value ≥ 8 kPa and FIB-4 index ≥ 1.35 at 48 weeks after SVR were significantly associated with any liver-related events in univariate analysis, with HRs (95% CI) of 1.08 (1.02-1.15), 5.68 (1.23-26.31), and 5.18 (1.10-24.32), p = 0.008, 0.026, and 0.037, respectively (Table 5).  Because the LS value and the FIB-4 index reflect the degree of liver fibrosis, we performed multivariate analyses of 3 models to identify independent factors associated with liver-related outcomes. When age and sex were controlled for, the LS value of ≥8 kPa but not the FIB-4 index at 48 weeks after SVR were significantly associated with liver-related events ( Table 6, Model 1 and 2). In the third model in which LS value and FIB-4 index were covariates, only the LS value remained independently associated with the occurrence of any liver-related events, with an adjusted HR (95% CI) of 5.04 (1.01-25.26), p = 0.049 (Table 6). Figure 1 depicts the cumulative incidence of any liver-related events based on the LS value at 48 weeks after SVR. liver-related outcomes. When age and sex were controlled for, the LS value of ≥8 kPa but not the FIB-4 index at 48 weeks after SVR were significantly associated with liver-related events ( Table 6, Model 1 and 2). In the third model in which LS value and FIB-4 index were covariates, only the LS value remained independently associated with the occurrence of any liver-related events, with an adjusted HR (95% CI) of 5.04 (1.01-25.26), p = 0.049 (Table 6). Figure 1 depicts the cumulative incidence of any liver-related events based on the LS value at 48 weeks after SVR.

Discussion
This study tracked the progression of liver fibrosis and identified the LS value as a predictor of liver-related events in HCV-infected patients who had received successful DAA therapy. Patients who achieved SVR after DAA therapy had a significant decrease in LS. Despite the regression of fibrosis, the incidence of unfavorable clinical outcomes, including HCC, after SVR remained. Accordingly, our findings emphasized the importance of monitoring LS after completion of DAA therapy, particularly at 48 ± 24 weeks, to determine the need for long-term follow-up for HCC surveillance and monitoring of liver decompensation.
The degree of liver fibrosis is known to be significantly reduced after achieving SVR [7,8,22]. In our study, patients who achieved SVR had a significant decrease in LS, with the greatest decrease of 3.5 pKa at 24 weeks, and the improvement of LS was modest at 72-96 weeks post-SVR. This finding was consistent with a previous study, which reported that patients who achieved SVR had significantly reduced LS values, with the greatest difference between baseline and 24 weeks, but no significant difference between 24 and 48 weeks [7]. A systematic review and meta-analysis of 24 studies which estimated the weighted mean difference in LS reported a significant decrease in LS of approximately 3.1 kPa from baseline to 6-48 weeks after achieving SVR, in contrast to an unchanged LS in those who did not achieve SVR [23].
Non-invasive tests have been used to longitudinally follow patients with HCV infection and to assess the effectiveness of antiviral treatment. A previous retrospective study found that the FIB-4 index improved significantly at 12 weeks after SVR in patients treated with a sofosbuvir-based regimen [24], with the mean ± SD of the FIB-4 index decreasing from 2.7 ± 2.2 to 2.0 ± 1.6 (p < 0.01). Another retrospective study of patients treated with paritaprevir/ritonavir/ombitasvir plus dasabuvir found that the median (IQR) of the FIB-4 index decreased from 3.6 (2.3-5.4) to 2.7 (2.0-3.8) at 12 weeks after completion of treatment (22). Similar to these studies, our cohort had a significantly lower FIB-4 index 48 weeks after successful DAA treatment.
Despite a significant reduction in the LS value, our study found that the incidence rates of all liver-related events and HCC were 2.36 per 100 person-years, and 1.17 per 100 patientyears, respectively, during the mean follow-up duration of 41 months, or 2.7 years, after SVR. All patients who developed HCC after SVR had a decrease in LS between baseline and 48 weeks after SVR, and 5 of 6 patients had cirrhosis prior to treatment. Previous studies reported similar HCC incidence rates ranging from 0.90 to 1.5 per 100 patientyears, with a higher HCC incidence rate in patients with cirrhosis who achieved SVR versus those without cirrhosis [12,13,25]. Another study reported the incidence of HCC, all-cause mortality, and liver-related mortality of 5.6%, 2.4%, and 10.4% in 125 patients who had SVR after HCV treatment [8]. No liver-related mortality was detected in our cohort, possibly due to the relatively short follow-up period. As reported in a long-term follow-up study of patients with HCV-related cirrhosis, the risk of HCC, liver decompensation, and liver-related mortality persisted for up to 8 years after SVR [26].
Most importantly, we found that an LS value of ≥8 kPa at 48 weeks post-SVR was significantly associated with poor liver-related outcomes. LS <8 kPa is currently recommended as the cut-off for ruling out advanced fibrosis in patients with alcoholic liver disease and non-alcoholic fatty liver disease [19]. However, due to inconsistencies in previous studies' results, using LS cut-off values by TE for determining fibrosis regression after SVR in HCV-infected patients has not been recommended [19]. In a study which compared preand post-treatment LS measured by TE and liver biopsy, the LS cut-off at 12 kPa was found to be suboptimal with 95% specificity but 61% sensitivity for diagnosing cirrhosis after SVR [27]. By contrast, another study in patients with recurrent HCV infection after liver transplantation demonstrated that LS by TE one year post-SVR accurately predicted the presence of advanced fibrosis, with the best cut-off values at 10.6 kPa and 14 kPa, respectively, to rule out and rule in advanced fibrosis [28]. Moreover, many studies attempted to identify the relationship between LS (before, after and changing after DAA treatment) and the development of HCC. Table 7 summarizes observational studies that assessed changes in LS after SVR and their association with the incidence of HCC in chronic HCV-infected patients. Most comparable to our study, a study in a population of compensated advanced chronic liver disease patients who achieved SVR after DAA reported that LS < 10 kPa at follow-up was a predictor of a lower HCC incidence rate at <1 per 100 patient-years [12]. As suggested by the guideline, the significant LS decrease observed after SVR highlights the need for lower cut-off values to be defined and validated [19]. Thus, our study provides an LS value that could be further investigated as a cut-off for predicting poor long-term liver-related outcomes in chronic HCV-infected patients following SVR.  It remains controversial whether the baseline LS value prior to DAA therapy and the degree of LS improvement is related to the progression of liver fibrosis or HCC occurrence after SVR. This study revealed that neither the baseline LS value nor the degree of LS improvement was related to the poor outcomes. Similarly, previous prospective studies found no significant differences in HCC incidence among patients with a baseline LS ≥ 20 kPa and LS < 20 kPa, and that a 20% reduction in LS was not associated with a reduced risk of HCC and liver-related events [12,32]. By contrast, other studies reported a high LS value before DAA initiation was a risk factor for HCC [29,31]. A baseline LS of >21.3 kPa was found to be associated with a 4.2-fold increased risk of HCC [29]. Another study reported that the three-year estimated incidence of de novo HCC was 20% in patients with baseline LS ≥30 kPa compared to only 5% in patients with LS ≤ 30 kPa [31]. A retrospective study also reported that a 30% improvement in LS was inversely associated with the risk of HCC, but this association was not observed in our cohort [30]. The discrepancies in results could be explained by the differences in participant selection, with varying degrees of liver fibrosis and duration of the follow-up period.
Regarding liver steatosis, our study found no significant changes in CAP between baseline and any time points following SVR. This finding was in line with a previous study reporting that liver steatosis did not significantly improve after treatment and that the prevalence of hepatic steatosis in chronic HCV-infected patients after SVR with DAA was approximately 47% [33].
The current study illustrates the importance of monitoring LS after DAA therapy completion to predict the need for long-term follow-up for liver-related complications. However, the results of the study should be interpreted in light of its limitations. First, due to the retrospective design, each patient did not have LS measurement at every time point after SVR; therefore, comparing each pair of time points was not possible. Second, the data on alcohol consumption were incomplete or missing in the medical records of most patients; thus, the effect of alcohol use on the occurrence of liver-related events could not be evaluated. Third, the number of patients included in the study was relatively small, which potentially carries the risk of overestimating the magnitude of an association [34]. Nonetheless, the observed association was consistent with the majority of previous studies. Fourth, given the relatively short follow-up period, it was possible that some clinical events would have occurred later. Lastly, the number of patients with newly diagnosed HCC was insufficient to investigate the relationship between LS after SVR and HCC development. Further studies with more patients and a longer follow-up time are warranted to gain a better understanding of the liver-related outcomes caused by changes in LS after DAA therapy.

Conclusions
In summary, chronic hepatitis C patients who achieve SVR with DAA therapy have a significant improvement in LS. At 48 weeks after SVR, LS ≥ 8 kPa is significantly associated with the occurrence of liver-related events. Those HCV-infected patients who have LS ≥ 8 kPa at 48 weeks post-SVR should be closely monitored and given measures to prevent and detect liver-related complications.