Efficacy and safety of glecaprevir/pibrentasvir in patients with chronic HCV infection and psychiatric disorders: An integrated analysis

Abstract Although direct‐acting antivirals (DAAs) for chronic hepatitis C virus (HCV) infection are highly efficacious and safe, treatment initiation is often limited in patients with neuropsychiatric disorders due to concerns over reduced treatment adherence and drug–drug interactions. Here, we report adherence, efficacy, safety and patient‐reported outcomes (PROs) from an integrated analysis of registrational studies using the pangenotypic DAA regimen of glecaprevir and pibrentasvir (G/P). Patients with chronic HCV genotypes 1‐6 infection with compensated liver disease (with or without cirrhosis) receiving G/P for 8, 12 or 16 weeks were included in this analysis. Patients were classified as having a psychiatric disorder based on medical history and/or co‐medications. Primary analyses assessed treatment adherence, efficacy (sustained virologic response at post‐treatment week 12; SVR12), safety and PROs. Among 2522 patients receiving G/P, 789 (31%) had a psychiatric disorder with the most common diagnoses being depression (64%; 506/789) and anxiety disorders (27%; 216/789). Treatment adherence was comparably high (>95%) in patients with and without psychiatric disorders. SVR12 rates were 97.3% (768/789; 95% CI = 96.2‐98.5) and 97.5% (1689/1733; 95% CI = 96.7‐98.2) in patients with and without psychiatric disorders, respectively. Among patients with psychiatric disorders, SVR12 rates remained >96% by individual psychiatric diagnoses and co‐medication classes. Overall, most adverse events (AEs) were mild‐to‐moderate in severity with serious AEs and AEs leading to G/P discontinuation occurring at similarly low rates in both patient populations. In conclusion, G/P treatment was highly efficacious, well‐tolerated and demonstrated high adherence rates in patients with chronic HCV infection and psychiatric disorders.


| INTRODUC TI ON
Chronic hepatitis C virus infection is associated with neuropsychiatric disorders in up to 50% of cases. 1,2 Historically, patients with comorbid psychiatric disorders were less likely to receive HCV treatment since interferon (IFN)-based regimens can induce depression and other neuropsychiatric manifestations including insomnia, irritability and mood changes. 3 Treatment adherence among this patient population was also a concern due to IFN's psychiatric side effect profile and perceived risks of lower adherence in patients with psychiatric and/or substance use disorders. 3 Yet, fatigue and psychological issues contribute significantly to quality-of-life (QoL) impairments in patients with chronic HCV infection, both of which can be improved by the achievement of sustained virologic response (SVR). 4,5 The introduction of direct-acting antivirals (DAAs) provided IFNfree treatment regimens that likely are more suitable for patients with chronic HCV infection and comorbid psychiatric disorders.
Both clinical trials and real-world evidence have demonstrated that these all-oral, IFN-free regimens are highly efficacious and well tolerated with minimal treatment-emergent neuropsychiatric side effects in patients with chronic HCV infection. [6][7][8] Based on these data, recent HCV treatment guidelines recommend DAA regimens without any restrictions based on psychiatric comorbidities. 9,10 However, there has been more limited use of DAAs in patients with psychiatric disorders in both late phase trials and clinical practice potentially due to concerns about treatment adherence and drug-drug interactions (DDIs) with neuropsychiatric co-medications. 3,[11][12][13][14] Thus, there is an unmet need to better understand the impact of psychiatric disorders on the treatment adherence, efficacy and safety of DAA regimens.
Glecaprevir (GLE; NS3/4A protease inhibitor identified by AbbVie and Enanta) and pibrentasvir (PIB; NS5A inhibitor) are potent pangenotypic inhibitors co-formulated as G/P, an all-oral, once-daily and pangenotypic DAA regimen that demonstrated high efficacy, and favourable safety and DDI profiles in patients with chronic HCV infection. In vitro, glecaprevir and pibrentasvir exhibited nanomolar and picomolar potencies, respectively, against all major HCV genotypes and both retained their activity against most resistance-associated substitutions. 15,16 Phase 1 trials investigated and thoroughly characterized the DDI profile of G/P, finding limited interactions and demonstrating that the majority of concomitant medications, including neuropsychiatric medications, can be safely taken with G/P without dose modification. 17,18 In late phase clinical trials, G/P was highly efficacious and safe in patients with chronic HCV genotypes 1-6 infection including in patients with compensated cirrhosis, end-stage renal disease and co-infection with human immunodeficiency virus (HIV). [19][20][21][22][23][24][25][26] Preliminary reports from real-world cohorts have supported these clinical trial findings with G/P showing similarly high effectiveness and a favourable safety profile in clinical practice. 27,28 Here, we present an integrated analysis of ten Phase 2 and Phase 3 studies aimed at evaluating the impact of psychiatric disorders on the treatment adherence, efficacy, safety and patient-reported outcomes (PROs) with G/P treatment.

| Analysis set
Data were pooled from 2,522 patients with chronic HCV genotype 1-6 infection and either without cirrhosis or with compensated cirrhosis who received G/P in ten Phase 2 and Phase 3 clinical trials that assessed efficacy, safety, treatment adherence and PROs (SURVEYOR-I and   SURVEYOR-II and MAGELLAN-1, and ENDURANCE-1, ENDURANCE-2,   ENDURANCE-3 and ENDURANCE-4, and EXPEDITION-1, EXPEDITION-2 and EXPEDITION-4). [19][20][21][22][23][24][25][26] This integrated analysis set included all patients who received at least one dose of glecaprevir 300 mg and pibrentasvir 120 mg either as separate tablets (Phase 2 formulation) or co-formulated tablets dosed orally as three pills for a total 300mg/120mg dose (Phase 3 formulation). Both formulations were given as once-daily, all-oral regimens for 8, 12 or 16 weeks. All authors had access to data, and reviewed and approved the final manuscript.

| Patients
Complete inclusion and exclusion criteria were similar across all clinical trials minus key trial-specific eligibility criteria noted in Table S1.
Adults (≥18 years of age) with chronic HCV genotype 1-6 infection were eligible for the studies if they were positive for anti-HCV antibody with a plasma HCV RNA viral load ≥10 000 IU/mL in Phase 2 engage in rigorous, independent scientific research, and will be provided following review and approval of a research proposal and Statistical Analysis Plan (SAP) and execution of a Data Sharing Agreement (DSA). Data requests can be submitted at any time and the data will be accessible for 12 months, with possible extensions considered. For more information on the process, or to submit a request, visit the following link: https ://www.abbvie.com/ our-scien ce/clini cal-trial s/clini cal-trialsdata-and-infor mation-shari ng/data-andinfor mation-shari ng-with-quali fied-resea rchers.html. events (AEs) were mild-to-moderate in severity with serious AEs and AEs leading to G/P discontinuation occurring at similarly low rates in both patient populations. In conclusion, G/P treatment was highly efficacious, well-tolerated and demonstrated high adherence rates in patients with chronic HCV infection and psychiatric disorders.  Adherence was defined as the lowest adherence to G/P being ≥80%

| Procedures
and ≤120% at all intervals between baseline and end of treatment. Safety was evaluated by monitoring adverse events (AEs), vital signs, physical examination findings, electrocardiography and clinical laboratory tests. Nonserious and serious AEs were monitored throughout G/P treatment until 30 days post-treatment and up to 24 weeks post-treatment, respectively. Any AE with an onset date after the first G/P dose and no more than 30 days after the last G/P dose was classified as a treatment-emergent adverse event (TEAE).
All AEs were coded using the Medical Dictionary for Regulatory Activities (MedDRA) and were assessed for their relationship with G/P by study investigators.

| Endpoints
The primary endpoints of this integrated analysis were the efficacy and safety of G/P in patients with chronic HCV genotype 1-6 infection with or without a psychiatric disorder. Efficacy was evaluated by SVR (HCV RNA < lower limit of quantification) at 12 weeks posttreatment in the intent-to-treat (ITT) population for patients with or without a psychiatric disorder. Safety was evaluated by characterizing reported AEs and the number and percentage of G/P-treated patients who reported treatment-emergent adverse events and laboratory abnormalities, both in total and stratified by the presence or absence of a psychiatric disorder. Additional endpoints included analyses of treatment adherence, patient-reported health outcomes related to neurocognitive function and subgroup analyses of SVR12 in patients by neuropsychiatric co-medications or psychiatric diagnoses.

| Statistical analysis
The number and percentage of patients in the ITT population achieving SVR12 for patients with or without psychiatric disorders were summarized with two-sided 95% confidence intervals calculated using the normal approximation to the binomial distribution. Further analyses evaluated SVR12 rates by ITT analysis in subgroups of patients with psychiatric disorders in order to assess whether common psychiatric disorders and/or co-medications affected achievement of SVR12.  Table 1 delineates the baseline and disease characteristics of patients with or without a psychiatric disorder. Patients with psychiatric disorders were more often female (49% vs 40%, respectively), white (87% vs 77%, respectively), GT3-infected (32% vs 24%, respectively) and had higher prevalence of cirrhosis (16% vs 11%, respectively) and medical history of injection drug use (56% vs 34%, respectively) compared to those without a psychiatric disorder.

| Baseline patient demographics and characteristics
The most common concomitant medication for each of these neuropsychiatric drug classes is listed in Table 2 for patients with or without psychiatric disorders when appropriate. Among the most common neuropsychiatric medications used concomitantly with G/P, the University of Liverpool website only identified quetiapine (N = 47), hydrocodone (N = 77) and oxycodone (N = 81) as neuropsychiatric co-medications with potential DDIs with G/P ( Table 2).

| Treatment adherence
Treatment adherence was similarly high among patients with or without a psychiatric disorder (95.4% and 96.7%, respectively; Table 1). Table 3 illustrates high (>94%) treatment adherence rates among patients with histories of depression or anxiety disorders; treatment adherence was lowest in patients with a history of bipolar disorder (89.5%). Treatment adherence remained high (>94%) for those prescribed neuropsychiatric co-medication classes (Table 3) and for patients taking >1 psychiatric co-medication or 16-week G/P treatment (Table S2). failures from premature discontinuations of G/P and lost to followup each occurred in less than 1% of patients regardless of the presence or absence of a psychiatric disorder.  . Similarly, SVR12 rates were >97% for all of the most common neuropsychiatric co-medication classes ( Figure 2) and remained high (>96%) for patients with psychiatric disorders not taking a concomitant psychiatric medication ( Figure S4).

| Safety outcomes
Among the 1697 (67%) patients experiencing an AE, 610 (77%) patients with a psychiatric disorder reported an AE compared to 1087 (63%) patients without a psychiatric disorder (Table 4). Most AEs were classified as either mild (Grade 1) or moderate (Grade 2) in severity for patients with (568/610; 93%) and without a psychiatric disorder (1035/1087; 95%). Overall, the most common AEs occurring in ≥10% of patients with or without a psychiatric disorder, respectively, were headache (20% vs 16%), fatigue (18% vs 13%) and nausea (13% vs 8%), and these tended to occur more often in patients with a psychiatric disorder. Neuropsychiatric AEs were not observed in ≥10% of patients for either population, but tended to occur more often in patients with a psychiatric disorder than those without a psychiatric disorder (8% vs 3%, respectively).
Adverse events leading to G/P discontinuation and G/P-related serious AEs were rare (<1%) in both patients with and without a psychiatric disorder ( at Day 5, respectively. Similar rates of AEs and G/P-related AEs leading to discontinuation were observed in patients taking a neuropsychiatric co-medication with a potential DDI, namely quetiapine, oxycodone and hydrocodone (Table S3). Although serious AEs were reported at numerically higher rates for patients taking oxycodone (11%) or hydrocodone (6%), none of the serious AEs were due to respiratory depression or related to G/P (Table S3).
Laboratory abnormalities were also rare (<1%) in patients with and without a psychiatric disorder (Table 4)  TA B L E 4 Adverse events and laboratory abnormalities post-treatment week 12. A numerically higher increase was observed in patients with a psychiatric disorder (3.6 ± 11.7) compared to those without a psychiatric disorder (2.2 ± 9.6). Likewise, available data from 542 and 1080 patients with and without psychiatric disorders, respectively, showed a trend at post-treatment week 12 towards slightly decreased fatigue scores as assessed by the FSS in patients with and without psychiatric disorders (−0.5 ± 1.6 and −0.3 ± 1.6, respectively).

| D ISCUSS I ON
The pangenotypic, once-daily DAA regimen of G/P achieved high (>97%) SVR12 rates in both patients with and without a psychiatric disorder. In patients with a psychiatric disorder, G/P achieved >96% SVR12 rates regardless of psychiatric diagnoses or the prescription of neuropsychiatric co-medications, in spite of a numerically higher prevalence of GT3 infection in patients with psychiatric disorders likely due to the higher prevalence of GT3 infection among patients with a history of substance use. 29,30 Overall, treatment adherence was >95% for both patients with or without psychiatric disorders. Adherence among patients with chronic HCV infection and comorbid psychiatric disorders remains a concern despite the advent of IFN-free DAA treatments. Lower adherence is hypothesized due to cases of severe, untreated psychiatric disorders and higher rates of comorbid substance use disorders 3,11-13 ; however, DAA regimens have shown similarly high adherence rates in clinical trials among patient populations traditionally considered to be at-risk for nonadherence including patients with psychiatric disorders, people who inject drugs and patients on stable opioid substitution therapy. 29,31 In its clinical trial programme, overall adherence to G/P regimens has been reported at >95% with comparable rates among patients with psychiatric disorders, people who inject drugs and patients on stable opioid substitution therapy. 33 Further long-term follow-up may be necessary in order to observe whether these trends are both robust and durable over time.
There are limitations to this analysis that are inherent to its design. Since this is a post hoc analysis, the comparisons between patients with and without a psychiatric disorder were not pre-specified and, thus, are not powered to assess for differences between these patient populations. It is not valid to compare psychiatric AEs between patients with or without a psychiatric disorder as there are most likely to be different risk factors or comorbidities for each of these groups contributing to their psychiatric AEs. The ideal analysis would involve a direct comparison of patients treated with or without G/P with psychiatric disorders. Additionally, the patients with psychiatric disorders enrolled in clinical trials are a more selected population potentially with more controlled disorders and an underrepresentation of Black or African-Americans compared to patients in the real-world setting; therefore, it is important to validate these findings with data obtained from real-world clinical practice. Of note, 30% of patients with psychiatric disorders were treatment-experienced with most receiving interferon-containing regimens, suggesting that these patients had less severe psychiatric disorders since interferon is contraindicated for patients with severe psychiatric disorders. Additionally, 77 patients were excluded from these studies as a result of a clinically significant abnormality other than HCV infection due to a cluster of clinical abnormalities of which may or may not have included an uncontrolled psychiatric disorder.
Full details are listed in the Supporting Information. Finally, treatment adherence was assessed in these clinical trials by measuring pill counts that provide quantitative data on adherence at specified visits, but can overestimate overall adherence and lacks any information on dose timing. 50 Overall, among the 789 (31%) patients with a psychiatric disorder in its clinical trials programme, G/P demonstrated comparably high SVR12 rates and treatment adherence to those observed in patients without a psychiatric disorder. Despite a higher rate of mild-to-moderate AEs in patients with psychiatric disorders, G/P was well tolerated with similarly low rates of serious AEs or AEs leading to G/P discontinuation in patients with or without psychiatric disorders.
Thus, G/P is a pangenotypic regimen suitable for treatment of the ≤50% of patients with chronic HCV infection and comorbid psychiatric disorders.

ACK N OWLED G EM ENT
Medical writing support was provided by Daniel O'Brien, Ph.D., of AbbVie.