Introduction
Hepatitis C virus (HCV) infection is one of the leading causes of cirrhosis and estimated to affect more than 185 million people worldwide.1 Direct-acting antiviral (DAA) therapy is the cornerstone of the treatment of chronic HCV infection. The vast majority of HCV-infected patients can be cured with current DAA treatment.2–5 It has been shown that eradication of HCV is associated with such benefits as decreased overall mortality, improved quality of life, and reduced healthcare utilization.6,7 Therefore, the goal of the treatment is to eradicate HCV RNA, predictable by attainment of a sustained virologic response (SVR; defined as undetectable of RNA level 12 weeks following the completion of DAA therapy).
Proton pump inhibitors (PPIs) are one of the most commonly prescribed medications worldwide for the treatment of all acid-related disorders, including gastroesophageal reflux disease and peptic ulcer. All DAA therapy can interact with PPIs, which affect gastric pH and can affect DAA bioavailability, thereby leading to sub-therapeutic levels of antiviral drugs and possibly to failure to achieve SVR.8–11 In fact, recent epidemiologic studies have reported the possible association between HCV-infected patients with DAA therapy concomitant use PPIs and lower odds of achieving SVR compared to non-PPI users.8,10–17 However, the results are still inconsistent. Thus, this systematic review and meta-analysis was conducted to summarize all available evidence with the aim of better characterizing this relationship.
Methods
Information sources and search strategy
A systematic literature search was conducted using the Embase and Medline databases from inception to March 2017 to identify all original studies that investigated the safety and efficacy of DAAs in patients with HCV infection taking PPIs versus those without PPIs. The systematic literature review was independently conducted by three investigators (K.W., S.C., and W.C.) using the search strategy that included the terms for “hepatitis C”, “direct-acting antivirals”, “sustained virologic response”, and “proton pump inhibitors” as described in Online Supplementary Data 1. A manual search for additional potentially relevant studies was also performed using references of the included articles. No language limitation was applied. This study was conducted in agreement with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (commonly known as PRISMA) statement which is provided as Online Supplementary Data 2.
Selection criteria
Eligible studies were required to be case-control, cross-sectional or cohort studies that had investigated the safety and efficacy of DAAs in patients with HCV infection taking PPIs. They must provide the effect estimates (odds ratios (ORs), relative risks (RRs), hazard ratios (HRs) or standardized incidence ratio (SIR)) with 95% confidence intervals (CIs). Inclusion was not restricted by study size. When more than one article using the same database/cohort was available, the study with the most comprehensive data/analyses was included.
Retrieved articles were independently reviewed for their eligibility by the same three investigators. Any discrepancy was resolved by conference with all investigators. Newcastle-Ottawa quality assessment scale was used to appraise the quality of study in three areas, including the recruitment of cases and controls, the comparability between the two groups and the ascertainment of the outcome of interest for cohort study and the exposure for case-control study.18
Data abstraction
A structured data collection form was used to extract the following data from each study: title of the study, publication year, name of the first author, year of the study, country where the study was conducted, number of participants, demographic data of participants, definition of PPIs use, type of DAA therapy, outcome measurement, adjusted effect estimates with 95% CIs and covariates that were adjusted in the multivariable analysis. To ensure accuracy, this data extraction process was independently performed by two investigators (KW and WC) and was reviewed by the senior investigator (WC).
Statistical analysis
Data analysis was performed using the Comprehensive Meta-Analysis Software (version 2.2.064; Biostat Inc). Adjusted point estimates from each study were combined by the generic inverse variance method of DerSimonian and Laird, which assigned the weight of each study in the pooled analysis based on its variance.19 In light of the high likelihood of between-study variance due to the different study populations, DAA therapy and definition of PPI use, therefore a random-effect model was used. Cochran’s Q test and I2 statistic were used to determine the between-study heterogeneity. A value of I2 of 0–25% represented insignificant heterogeneity, 26–50% represented low heterogeneity, 51–75% represented moderate heterogeneity, and more than 75% represented high heterogeneity.20 Meta-regression was performed to assess the effect of DAA regimens (sofosbuvir (SOF) based vs. non-SOF-based regimens) and the use of ribavirin (RBV) on the SVR using a random-effects meta-regression.21 Egger’s regression symmetry test was used to assess for publication bias. A p-value less than 0.05 was considered statistically significant for all analysis.
Results
Seven hundred and twenty potentially eligible articles were identified using our search strategy (251 articles from Medline and 469 articles from Embase). After the exclusion of 251 duplicate articles, 469 articles underwent title and abstract review. Four hundred and fifty-six articles were excluded at this stage since they were case reports, correspondences, review articles or interventional studies, leaving 13 articles for full-text review. Two of those were excluded after the full-length review as they did not report the outcome of interest, while one article was excluded since for being a descriptive study without comparative analysis. Ten studies met our eligibility criteria. However, two studies used the same database.11,22 To avoid duplication, we excluded one of those studies, and we decided to exclude the study that had been published as an abstract22 and later published as an original article (which was more comprehensive and had a larger number of participants).11 Therefore, nine cohort studies with 32,684 participants met the eligibility criteria.8,10–17 The literature retrieval, review, and selection process are shown in Fig. 1. The main features and quality assessment of the studies included in this meta-analysis are shown in Table 1.
Table 1.Characteristics of included studies
| Study | Shiffman et al.16 | Tapper et al.10 | Terrault et al.11 | Spoutz et al. (abstract)17 |
|
| Country | USA, France and Spain | USA | USA | USA |
| Study design | Cohort study | Cohort study | Cohort study | Cohort study |
| Year | 2016 | 2016 | 2016 | 2016 |
| Number of participants | 2053 | 887 | 1788 | 547 |
| Participants | Treatment-naive or peg-interferon/RBV treatment-experienced HCV genotype 1-infected patients with or without compensated cirrhosis received OBV/PTV/r and DSV ± weight-based RBV | Chronic HCV patients treated using any duration of LDV/SOF ± RBV | HCV genotype 1-infected patients treated with LDV/SOF ± RBV | HCV genotype 1 patients treated with LDV/SOF ± RBV |
| Direct-acting antiviral therapy | OBV/PTV/r and DSV ± RBV | LDV/SOF ± RBV | LDV/SOF ± RBV | LDV/SOF ± RBV |
| Definition of PPI use | Concomitant PPI use (omeprazole, esomeprazole, dexlansoprazole, lansoprazole, pantoprazole, rabeprazole) | PPI use during HCV treatment as defined by a filled PPI prescription during therapy | Baseline PPI use | Acid-reducing therapy (PPI or H2RA) |
| Outcome | SVR12
PPI group: 95.1% (293/308, 95% CI: 92.1–97.0)
Control: 96.4% (1683/1745, 95% CI: 95.5–97.2) | SVR12
Any PPI use
PPI group: 97.8% (343/351, 95% CI: 96.4–99.2)
Control: 97.2% (345/355, 95% CI: 95.7–98.7)
High-dose PPI use
PPI group: 98% (146/149, 95% CI: 95.7–100)
Control: 97.2% (345/355, 95% CI: 95.7–98.7) | SVR12
PPI use
PPI group: 93.5% (472/505, 95% CI: 90.9–95.5)
Control: 97.2% (1247/1283, 95% CI: 96.1–98)
High-dose PPI use
PPI group: 92.3% (168/182, 95% CI: 87.4–95.7)
Control: 93.8% (242/258, 95% CI: 90.1–96.4) | SVR12
Acid-reducing therapy group: 92.3% (144/156)
Control: 94.1% (368/391) |
| Adjusted OR of lower SVR | 0.95 (0.52–1.80) | Any PPI use
0.34 (0.05–2.29)
High-dose PPI use
0.95 (0.37–2.47) | PPI use
0.41 (0.25–0.67)
High-dose PPI use
0.15 (0.05–0.47) | 0.59 (0.25–1.38) |
| Confounder adjustment | Age, sex, race, BMI, weight, HCV RNA, ALT, HCV genotype 1 subtype, IL28B genotype, prior HCV treatment status, cirrhosis, treatment regimen, geographic region, geographic region, history of diabetes, depression or bipolar disorder, bleeding disorders, and former injection drug use | Propensity score matching and adjusting for age, sex, ethnicity, practice type, treatment experience, genotype group, baseline viral load, the presence of cirrhosis, platelet count, and duration of treatment | Age, sex, HCV genotype, albumin, platelet count, total bilirubin, hemoglobin, baseline HCV RNA, cirrhosis status, history of antiviral treatment, history of hepatic decompensation, BMI | Yes, not specified |
| Quality assessment by Newcastle-Ottawa scale | Selection: 4
Comparability: 2
Outcome: 3 | Selection: 4
Comparability: 2
Outcome: 3 | Selection: 4
Comparability: 2
Outcome: 3 | Selection: 3
Comparability: 1
Outcome: 3 |
We found that the use of PPIs concomitant with DAAs among HCV-infected patients was associated with lower odds of achieving SVR compared with non-PPI users (pooled OR of 0.74, 95% CI: 0.63–0.88, p < 0.001), as shown in Fig. 2. The heterogeneity between studies of the overall analysis was insignificant, with an I2 of 24%. Subgroup analysis to address the association between PPI uses and SVR12 also demonstrated the association of PPI users with lower odds of achieving SVR12 compared with those with no use of PPI (pooled OR of 0.68, 95% CI: 0.51–0.9, p = 0.01), as shown in Fig. 3. The heterogeneity between studies of the overall analysis was low, with an I2 of 33%. Moreover, subgroup analysis restricted to only the studies with adjusted confounding factors also showed an association between PPI uses with lower odds of achieving SVR compared with non-PPI users (pooled OR of 0.66, 95% CI: 0.47–0.94, p = 0.02), as shown in Fig. 4. The heterogeneity between studies of the overall analysis was low, with an I2 of 50%.
Meta-regression analysis
Meta-regression showed no significant impact of the uses of SOF (p = 0.16) or RBV (p = 0.18) in DAA regimens on the association between PPI use and lower odds of achieving SVR12.
Evaluation for publication bias
There was no publication bias for the overall included studies, as assessed by funnel plotting (Fig. 5) and the Egger’s regression asymmetry test (p = 0.24) of the association between PPI use and lower odds of achieving SVR.
Discussion
To the best of our knowledge, this is the first systematic review and meta-analysis that summarized all available studies that have reported on the efficacy and safety of DAA therapy in HCV-infected patients taking PPIs. DAA therapies in combination with PPIs may result in increasing risk of failure to achieve SVR. We found an approximately 1.4-fold increased risk of failure to achieve SVR12 within the PPI users group compared to the non-PPI users group.
The true pathogenesis of this association is still unclear; however, there is potential explanation. Ledipasvir’s solubility decreases as gastric pH increases. Thus, acid reducing agents can affect drug absorption and drug level.23,24 The package label for ledipasvir (LDV) recommended that patients who take PPIs should not take a dose higher than omeprazole 20 mg daily or equivalent and taken fasting at the same time as LDV/SOF. Whether the patients can follow these recommendations is unknown and may affect the result of the real world data and this meta-analysis. Seven of nine studies in this meta-analysis reported the interaction of PPIs and the LDV/SOF regimen.8,10–14,17 Only one study by Terrault et al.11 showed a significantly decreased achievement of SVR among PPI users daily. Tapper et al.10 showed that twice daily PPI use was associated with lower odds ratio for SVR but not daily PPI use. This can imply that PPI use decreased odds of achieving SVR, especially twice daily PPI usage.
The systematic literature review process of this study was comprehensive, and the quality of included studies was good even though some of the included studies are abstracts. Moreover, the statistical heterogeneity of this meta-analysis was low. We acknowledge, however, that this study had some limitations and, thus, the results should be interpreted with caution. First, not all included studies used pharmacy records to confirm which patients filled their PPI prescriptions throughout the treatment course, as well as the dose, quantity of the pills dispensed, and the frequency. The method to define and measure PPI use by pharmacy records is better than that for data from PPI use at baseline. Thus, we may not know the data on the dose, frequency and refilled prescription of PPIs throughout the course of the treatment in most of the included studies. Besides, most of the included studies were abstracts and not yet published in full original studies. Therefore, the final data and report of each study may change, such as the number of participants and adjusted confounding factors analysis. However, we believe that the primary outcome of each study which focused on the efficacy and safety of PPI use among HCV-infected patients with DAA therapy will not change from the report of the published abstracts.
In summary, this study demonstrated a significantly increased risk of failure of achievement of SVR in HCV-infected patients taking DAA with PPIs compared to non-PPI users. Providers should consider whether PPI therapy is indicated for these patients and withdraw PPI therapy in the absence of indications.
Supplementary information
Supplementary Data 1
Search Strategy
(DOC)
Supplementary Data 2
PRISMA 2009 Checklist
(DOC)
Abbreviations
- BMI:
body mass index
- CI:
confidence interval
- DAA:
direct-acting antiviral
- DSV:
dasabuvir
- EBR:
elbasvir
- GZR:
grazopravir
- HCV:
hepatitis C virus
- HR:
hazard ratio
- IL28B:
interleukin-28B gene
- kPa:
kilopascal
- LDV:
ledipasvir
- mg:
milligram
- OBV:
ombitasvir
- OR:
odds ratio
- PPI:
proton pump inhibitor
- PTV/r:
paritaprevir/ritonavir
- RBV:
ribavirin
- RNA:
ribonucleic acid
- RR:
relative risk
- SIR:
standardized incidence ratio
- SOF:
sofosbuvir
- SVR:
sustained virologic response
Declarations
Conflict of interest
The authors have no conflict of interests related to this publication.
Authors’ contributions
Contributed to conception and design of the study, and critical revision (KW, WC), acquisition of data, analysis and interpretation of data (KW, SC, CT, VJ), manuscript writing (KW, SC), revised the article (CT, VJ). All authors approved the final revision.