Initial uptake, time to treatment, and real-world effectiveness of all-oral direct-acting antivirals for hepatitis C virus infection in the United States: A retrospective cohort analysis

Background Data on initiation and utilization of direct-acting antiviral therapies for hepatitis C virus infection in the United States are limited. This study evaluated treatment initiation, time to treatment, and real-world effectiveness of direct-acting antiviral therapy in individuals with hepatitis C virus infection treated during the first 2 years of availability of all-oral direct-acting antiviral therapies. Methods A retrospective cohort analysis was undertaken using electronic medical records and chart review abstraction of hepatitis C virus-infected individuals aged >18 years diagnosed with chronic hepatitis C virus infection between January 1, 2014, and December 31, 2015 from the Indiana University Health database. Results Eight hundred thirty people initiated direct-acting antiviral therapy during the 2-year observation window. The estimated incidence of treatment initiation was 8.8%±0.34% at the end of year 1 and 15.0%±0.5% at the end of year 2. Median time to initiating therapy was 300 days. Using a Cox regression analysis, positive predictors of treatment initiation included age (hazard ratio, 1.008), prior hepatitis C virus treatment (1.74), cirrhosis (2.64), and history of liver transplant (1.5). History of drug abuse (0.43), high baseline alanine aminotransferase levels (0.79), hepatitis B virus infection (0.41), and self-pay (0.39) were negatively associated with treatment initiation. In the evaluable population (n = 423), 83.9% (95% confidence interval, 80.1–87.3%) of people achieved sustained virologic response. Conclusion In the early years of the direct-acting antiviral era, <10% of people diagnosed with chronic hepatitis C virus infection received direct-acting antiviral treatment; median time to treatment initiation was 300 days. Future analyses should evaluate time to treatment initiation among those with less advanced fibrosis.


Introduction
Hepatitis C virus (HCV) infection is the most common blood-borne infection in United States, with approximately 2.2 to 3.8 million persons infected nationwide [1]. Left untreated, HCV infection can cause cirrhosis or hepatocellular carcinoma, and may lead to liver transplant or death [2]. Before 2011, interferon-based therapies were the main treatment for HCV infection, yielding sustained virologic response (SVR, or cure) rates of approximately 45% to 50% [3]. The introduction of first-generation direct-antiviral agents (DAA), administered with pegylated interferon and ribavirin, improved SVR rates to approximately 50% to 75% in treatmentnaive individuals with genotype (GT) 1 infection [4][5][6]. However, the improvements in SVR attained with first-generation agents were associated with an increased frequency of adverse events. In 2013, simeprevir and sofosbuvir were approved for the treatment of HCV infection [7]. Although not approved as a combination regimen, the off-label use of these agents led to the first all-oral DAA therapy for HCV GT1 infection. In addition, sofosbuvir plus ribavirin was approved for the treatment of GT2 and GT3 infection [8]. Subsequently, ledipasvir/sofosbuvir and ombitasvir, paritaprevir, and dasabuvir were approved in the United States and Europe. These agents offered short-duration, interferon-free treatments that could be administered with or without ribavirin and achieved SVR rates of >90% [9][10][11][12][13][14].
Additional barriers to the real-world treatment of HCV infection exist that are often controlled for in clinical trials. Real-world treatment of HCV infection may be complicated by comorbidities and concomitant medications, and economic considerations can prevent the successful initiation or completion of DAA-based therapy [15][16][17][18]. To date, data are limited concerning the proportion of HCV-infected individuals who are eligible for DAA treatment or who have initiated treatment, and few reports have evaluated the time to treatment initiation following diagnosis. The objectives of this study were to evaluate treatment initiation rates, time to treatment, and predictors of initiating treatment in individuals receiving all-oral DAA therapy in a real-world setting. Real-world effectiveness of all-oral DAA regimens and early discontinuation rates were also assessed.

Methods
This was a retrospective cohort analysis using electronic medical records (EMRs) and chart review abstraction of HCV-infected individuals from the Indiana University health database, which included data from 48,000 persons in 2015. The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines, and the protocol was approved by the Indiana University Office of Research Compliance institutional review board. Data regarding prior HCV treatment and comorbidities were captured through the Indiana Network for Patient Care, a health information exchange containing data from multiple Indiana healthcare institutions. Manual chart reviews were performed to validate HCV genotype, treatment start and end dates, and clinical characteristics including HCV RNA viral load, aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels, and platelet counts. A waiver of the informed consent process was approved by the appropriate institutional review board.

Study population
Individuals aged >18 years with a documented ICD-9 and/or ICD-10 code (International Classification of Diseases, 9 th or 10 th edition, Clinical Modification) of non-acute HCV infection and a documented encounter between January 1, 2014, and December 31, 2015 (the cohort identification period) were included. All participants with at least one record of an all-oral DAA prescription filled during the cohort identification period were included in the DAA-treated cohort. End-of-treatment for participants enrolled during the cohort identification period could extend to March 31, 2016. DAA regimens in this study included sofosbuvir/ ledipasvir ± ribavirin, paritaprevir/ritonavir/ombitasvir/dasabuvir (PrOD) ± ribavirin, sofosbuvir/simeprevir ± ribavirin, and sofosbuvir + ribavirin. Participants were required to have �6 months of historical records prior to the treatment index date, which was defined as the time at which an individual first received all-oral DAA treatment during the study period. Individuals receiving concurrent interferon therapy were excluded from the DAA-treated cohort.
HCV-infected individuals with no record of all-oral DAA treatment were included in the HCV untreated cohort.

Study measures
Baseline characteristics, prior HCV treatment, and HCV genotype were recorded. HCV genotype, fibrosis level, AST:platelet ratio index (APRI), intended treatment duration, and treatment completion duration were extracted from medical chart reviews. Fibrosis level was based on the APRI and Fibrosis 4 (FIB-4) indices. The FIB-4 score was calculated based on variables recorded before the DAA index date, using the following formula: APRI score was calculated via the following formula: Cirrhosis was defined as a FIB-4 score of >3.5. Co-morbidities such as hepatitis B virus infection, HIV, and history of kidney or liver transplant were defined by the presence of at least one ICD-9 CM or ICD-10 CM code during the baseline period of 1 year prior to and including the index date.

Study outcomes
Treatment initiation rates and time to treatment were calculated using the time from the first clinical encounter of HCV diagnosis to the first medication order of a DAA agent. Early treatment discontinuation was defined as an observed treatment duration minus treatment gap that was shorter than intended treatment duration by 14 days. SVR12 was defined as undetectable HCV RNA 12 weeks after the end of treatment, and end of treatment was calculated as the last date covered by prescription using medication dispensed information and number of days for which medication was supplied. SVR12 was calculated for the evaluable population (all participants with an available treatment outcome who completed the planned regimen or discontinued early) and the per-protocol population (all participants with an available treatment outcome who completed treatment).

Statistical analysis
Descriptive statistics and univariate analysis were used to describe the distribution of demographics and clinical and laboratory characteristics. Time to treatment was used to estimate the cumulative percentage of eligible HCV-infected individuals receiving DAA prescriptions. Individuals with no DAA prescription were assumed to have not received any treatment at Indiana University until the last encounter in the EMR, at which point the observed time to treatment was censored. Kaplan-Meier analysis was used to estimate the percentage of participants receiving prescriptions at any time from the first clinical encounter.
Univariate analysis was used to identify factors significantly associated with DAA prescription using t tests and chi-square tests. Cox regression analysis was used to predict factors associated with time to initiation of therapy. SAS (version 9.4, SAS Institute Inc., Cary, NC, USA) and Stata version 11 (Stata Corp., College Station, TX, USA) were utilized to conduct analyses.

Results
A total of 8,611 individuals with a diagnosis of chronic HCV, or a pharmacy order for a DAA agent between January 1, 2014, and December 31, 2015, were identified. After applying inclusion/exclusion criteria, 8,407 individuals with chronic HCV infection were included, of whom 830 initiated DAA therapy and 7,577 did not, resulting in a treatment initiation rate of 9.9%.

Time to treatment initiation and factors associated with treatment initiation
At year 1, the estimated incidence of treatment initiation was 8.80% ± 0.34%, and at year 2, this increased to 15% ± 0.5%. The median time to DAA initiation was 300 days among those who received treatment within the 2-year observation window (Fig 1), ranging from 25 to 675 days with a distribution as shown in S1 Fig. Sofosbuvir/ledipasvir ± ribavirin was the most commonly initiated treatment, comprising 382/830 (46%) of all treatments started, followed by sofosbuvir/ribavirin (25.7%) and sofosbuvir/simeprevir ± ribavirin (22.5%) (  (Table 3).
Manual chart review of a random sample from the untreated cohort (n = 287) suggests that the most common reasons for lack of initiation of therapy were absence of patient follow-up (33.7%) and insurance declined or other cost barrier (28.1%). Physician decision not to treat accounted for 16% of non-initiation cases. The reasons for not initiating DAA therapy are listed in S1 Table. Real-world effectiveness of all-oral direct-acting antivirals for hepatitis C virus infection Clinical characteristics and demographics of the treated cohort stratified by DAA regimen are reported in S2 Table. Ribavirin was added to sofosbuvir/ledipasvir most commonly for individuals with more advanced fibrosis, including compensated and decompensated cirrhosis, those with prior liver transplant, and those previously treated with an interferon-based therapy.
The treatment discontinuation rate was 4.2%, ranging from 2.3% in people receiving sofosbuvir/ledipasvir to 6.6% in those receiving sofosbuvir/ribavirin and 10% in those receiving PrOD ± ribavirin.

Discussion
All-oral DAA therapy has revolutionized the treatment of chronic HCV infection. Before the availability of DAA agents, SVR rates of up to 75% were observed in HCV GT1-infected individuals; however, side effects with interferon-based treatments were often severe, and many individuals were not eligible for interferon therapy. SVR rates of >95% have been reported with all-oral DAA combinations in individuals with GT1 infection, as confirmed in multiple observational cohort studies [19][20][21]. However, the benefits of highly efficacious therapies may be limited if these treatments cannot be accessed by those who require them. To date, data Real-world effectiveness of all-oral direct-acting antivirals for hepatitis C virus infection regarding the frequency of treatment initiation in the era of DAA therapy are limited. This study therefore serves as a reference for newer DAA therapies with improved access to care. The demographics of this study population mirror those of other HCV infection cohorts reported in registration trials, with most participants being white males [2,8,9,11]. The overall uptake of therapy among participants eligible for treatment during the first 2 years of the DAA era was low at 9.9%, and was similar to that previously reported with interferon-based treatments [22]. We observed an increase in the incidence of treatment initiation during the second year of the study; however, a median time to initiation of therapy of 300 days suggests that substantial improvements in overcoming barriers to treatment are still required. Insurance type was a significant predictor of treatment initiation, with the highest rates seen in those with Medicare and private insurance. Disease severity was also a predictor of initiation of therapy, with higher rates in those with clinical features suggesting advanced liver disease, including compensated and decompensated cirrhosis, greater Charlson comorbidity scores, and a history of liver transplant. It is likely that the higher rates of treatment initiation in these groups can be explained partially by their requirement for continuous medical care. Prior treatment was also correlated with initiation of treatment, whereas a history of drug abuse was Real-world effectiveness of all-oral direct-acting antivirals for hepatitis C virus infection negatively correlated with treatment initiation. The characteristics of our treatment cohort mirror the population initially prioritized for treatment in guidelines and by private and public insurers. Future studies should examine fibrosis status among those initiating therapy to determine if those with less severe disease and fewer comorbidities are also starting treatment. Current guidelines emphasize that all individuals with HCV infection should receive DAA therapy if they comply with the intended treatment [23]. More recent data suggest that individuals with depression, illicit substance use, and other psychiatric disorders can be successfully treated, with SVR rates similar to those without these comorbidities. It is also expected that future studies will demonstrate that a history of drug abuse does not negatively correlate with initiation of therapy [24,25]. In our study, the highest rates of treatment initiation were 39% and 36.3% in people with Medicare and private insurance, respectively, compared with only 15.9% of those with Medicaid. However, in several states treatment restrictions are gradually being reduced, which, given the sizeable population of HCV-infected individuals with Medicaid as their primary insurance, should improve DAA therapy initiation rates.
The most common intended treatment duration was 12 weeks with sofosbuvir/ledipasvir, with an overall SVR of 92.2% (n = 153/166). In people receiving sofosbuvir/ledipasvir, SVR rates were lower in cirrhotic than in non-cirrhotic individuals (88.9% vs. 94.2%). Those who received sofosbuvir/ledipasvir + ribavirin had an overall response rate of 84.6%, with no difference between treatment-naive and treatment-experienced individuals (84.2% vs. 85.2%). Other reports confirm that the real-world effectiveness of DAA agents varies by HCV population. HCV-TARGET is a prospective, longitudinal, observational study of patients with chronic HCV infection at academic and community centers from the United States, Canada, Germany, and Israel. This study has reported SVR rates of 84% in people with HCV GT1 infection receiving sofosbuvir/simeprevir for up to 16 weeks [19]; 94% in people with HCV GT1 infection receiving sofosbuvir/ledipasvir for 12 or 24 weeks [26]; and 93% in individuals with HCV GT1 infection receiving sofosbuvir/ledipasvir for 8 to 16 weeks [20]. In addition, studies from the Veterans Affairs population in the United States and a large US commercially insured population evaluating patients with HCV GT1 infection confirm SVR rates of 90% to 94% in individuals receiving sofosbuvir/ledipasvir [27][28][29][30].
An SVR rate of 79.5% was noted in those who received sofosbuvir/simeprevir with or without ribavirin. The combination of sofosbuvir/simeprevir is approved for the treatment of GT1-infected individuals [31]; however, in 2014 this was an off-label combination that represented the most efficacious regimen available for people with GT1 infection. In other realworld settings, rates of SVR12 in people receiving sofosbuvir/simeprevir were 64% in those with GT1a infection and 84% in those with GT1b infection [32]. SVR rates of 72% in those with GT2 infection receiving sofosbuvir/ribavirin and 35% in those with GT3 infection receiving sofosbuvir/ribavirin are also reported [32].
Early discontinuation rates were low (4.2%), ranging from 2.3% in participants receiving sofosbuvir/ledipasvir without ribavirin to 10% in those receiving PrOD ± ribavirin. These discontinuation rates are substantially lower than those reported with interferon-based therapy, likely owing to the dramatically different tolerability profiles between DAA and interferonbased treatments [3]. It is also apparent that discontinuation rates in the present study were higher among participants receiving a ribavirin-containing regimen, possibly because of the well-described tolerability profile of ribavirin. Welzel et al. also reported discontinuation rates of 6.7% in people receiving sofosbuvir/ribavirin in a real-world setting; however, other studies indicate that discontinuation rates with DAA therapies are similar (approximately 1% to 3%) regardless of ribavirin use [19,20].
This study had several limitations. The source of the database was a large center in Indiana; thus, the findings may not be generalizable to the general US population. The study period included the first 2 years that all-oral DAA therapies were available; since this time, both guidelines and insurers have broadened the criteria for treatment eligibility, with the present American Association for the Study of Liver Diseases guidelines stating that all people with HCV infection without a limited lifespan and who can comply with treatment should be considered for therapy. Since these data were collected, there have been substantial changes in the use of DAA therapies for the treatment of HCV infection; these data are therefore unlikely to accurately represent the current utilization of DAA treatments. The use of EMRs may limit the granularity or the completeness of our data. We therefore implemented a manual chart review to minimize the impact of missing values and to validate laboratory results.

Conclusion
Data from the present study provide insight into initiation of DAA therapy during the early years of the availability of all-oral regimens. We found that treatment was initiated in 9.9% of HCV-infected individuals and time to initiation was 300 days. Expanding treatment eligibility criteria and reducing economic barriers should increase rates of treatment initiation and reduce time to treatment. Future studies, based on more recent treatment periods, are likely to demonstrate improved access to treatment and broadened treatment eligibility including individuals with less advanced disease.