Nicole Loo, Eric Lawitz, Naim Alkhouri, Jennifer Wells, Carmen Landaverde, Angie Coste, Rossalynn Salcido,Michael Scott, Fred Poordad
Abstrac t BACKGROUND The hepatitis C virus (HCV) NS5A inhibitor ABT-267 (ombitasvir, OBV), the HCV NS4/4A protease inhibitor ABT-450 (paritaprevir, PTV), the CYP3A inhibitor ritonavir (r) and the non-nucleoside NS5B polymerase inhibitor ABT-333 (dasabuvir, DSV) (OBV/PTV/r + DSV) with or without ribavirin (RBV) is a direct-acting antiviral regimen approved in the United States and other major countries for the treatment of HCV in genotype 1 (GT1) infected patients. Patients with HCV who are considered “hard-to-cure” have generally been excluded from registration trials due to rigorous study inclusion criteria, presence of comorbidities and previous treatment failures.AIM To investigate the efficacy of this regimen in HCV G1-infected patients historically excluded from clinical trials.METHODS Patients were ≥ 18 years old and chronically infected with HCV GT1 (GT1a, GT1b or GT1a/1b). Patients were treatment-naïve or previously failed a regimen including pegylated interferon/RBV +/- telaprevir, boceprevir, or simeprevir.One hundred patients were treated with the study drug regimen, which was administered for 12 or 24 wk +/- RBV according to GT1 subtype and presence/absence of cirrhosis. Patients were evaluated every 4 wk from treatment day 1 and at 4 and 12 wk after end-of-treatment.RESULTS Many of the patients studied had comorbidities (44.2% hypertensive, 33.7%obese, 20.2% cirrhotic) and 16% previously failed HCV treatment. Ninety-six patients completed study follow-up and 99% achieved 12-wk sustained virologic response. The majority (88.4%) of patients had undetectable HCV RNA by week 4. The most common adverse events were fatigue (12%), headache (10%),insomnia (9%) and diarrhea (8%); none led to treatment discontinuation. Physical and mental patient reported outcomes scores significantly improved after treatment. Almost all (98%) patients were treatment compliant.CONCLUSION In an all-comers HCV GT1 population, 12 or 24-wk of OBV/PTV/r + DSV +/-RBV is highly effective and tolerable and results in better mental and physical health following treatment.
Key words: Hepatitis C; Ombitasvir; Paritaprevir; Ritonavir; Dasabuvir; Genotype 1 outside the submitted w ork; Dr.Loo reports grants from AbbVie during the conduct of the study;personal fees from Gilead outside the submitted work; Ms. Coste, Dr.Landaverde, Ms. Salcido, Dr. Scott and Dr. Wells report grants from AbbVie during the conduct of the study.
The National Health and Nutrition Εxamination Survey (NHANΕS) approximates that 3.2 million hepatitis C virus (HCV) patients are living in the United States[1].Among these patients, most will develop chronic infection that can ultimately lead to hepatic fibrosis and cirrhosis[2]. Cirrhotic HCV patients have an increased risk of developing d ecompensated cirrhosis, hepatocellular carcinoma (HCC)[3,4]. and commonly require liver transplantation[5].
The recent era of improved treatment regimens for HCV has contributed to much better outcomes in these patients[6], with the goal of treatment being to eradicate the virus and prevent the development of cirrhosis and its complications. Successful treatment of HCV is defined in terms of sustained virologic response (SVR), which is undetectable levels of HCV viral RNA in the blood 12 w k after completion of therapy[7]. Currently, treatment for HCV involves the use of direct-acting antiviral(DAA) regimens, which are molecules that target specific nonstructural proteins of the virus and disrupt viral replication and infection. One such DAA regimen is the HCV NS5A inhibitor ombitasvir (OBV), HCV NS4/4A protease inhibitor paritaprevir(PTV), the CYP3A inhibitor ritonavir and the non-nucleoside NS5B polymerase inhibitor dasabuvir (DSV) (OBV/PTV/r + DSV) with or without ribavirin (RBV). In registration trials for this regimen, SVR rates reached as high as 99% in HCV genotype 1 (GT1) patients[8].
Patients w ith HCV w ho are consid ered “hard-to-cure” have generally been excluded from registration trials due to rigorous study inclusion criteria, presence of comorbidities and previous treatment failures. Here we report the results of a phase 4,open label study that evaluated the safety and efficacy of OBV/PTV/r + DSV +/-RBV in a real-world clinical setting in patients who have historically been excluded from clinical trials. We also report on the patient quality of life, dosing adherence and whether resistance-associated substitutions (RASs) impact achievement of SVR.
All p atients p rovid ed w ritten informed consent. The stud y w as cond ucted in accord ance w ith the International Conference on Harmonisation, ap p licable regulations, and the princip les of the Declaration of Helsinki. Upon signing the informed consent, patients w ere consid ered enrolled and screening and baseline assessments w ere initiated. Patients w ere screened from April 21, 2016 through December 5, 2016. Screening p roced ures at baseline includ ed a med ical history,physical exam, blood chemistries, HCV RNA PCR and HCV genotype assessments.
Εligible patients were ≥ 18 years old and chronically infected with HCV GT1 (GT1a,GT1b or GT1a/1b). Patients w ere either treatment naïve or p reviously failed a regimen includ ing pegylated interferon (p egIFN)/RBV w ith or w ithout telaprevir,bocep revir, or simeprevir. Patients w ere exclud ed from the stud y if they w ere currently taking or p lanning on taking any med ications contraind icated in the OBV/PTV/r + DSV p ackage insert, had evid ence of d ecomp ensated liver disease(Child-Pugh B or C, including the presence of clinical ascites, bleed ing varices, or hepatic encep halop athy), hemoglobin < 8 g/d L, platelets < 25000 cells/mm3, all nucleated cells < 500 cells/mm3, bilirubin > 3, international normalized ratio ≥ 2.3,serum albumin < 2.8 and glomerular filtration rate < 30 m L, alanine aminotransferase or aspartate aminotransferase > 10 × ULN, consume > 3 alcoholic drinks daily or have uncontrolled HIV or hep atitis B virus coinfection. Patients w ith cirrhosis w ere required to have serum alpha fetoprotein < 100 ng/mL and imaging ruling out HCC within 3 mo of screening visit.
Treatment w as initiated w hen screening and baseline procedures w ere completed,and p atient eligibility w as confirmed. Patients received the US p ackage insert recommend ed d ose: Tw o OBV/PTV/r 12.5/75/50 mg tablets once d aily (in the morning) and one DSV 250 mg tablet tw ice daily (morning and evening) with a meal w ithout regard to fat or calorie content. RBV w as given to some patients based on approved US package insert and was dosed at 1000 mg/d (< 75 kg) or 1200 mg/d (≥75 kg), d ivid ed and ad ministered tw ice-daily w ith food. Duration of treatment w as determined based on GT1 subtype and the presence or absence of cirrhosis. GT1a patients w ith compensated cirrhosis w ere treated for 24 w k includ ing RBV, GT1a patients w ithout cirrhosis and GT1b patients with compensated cirrhosis were treated for 12 w k includ ing RBV, and GT1b patients w ithout compensated cirrhosis w ere treated for 12 wk w ithout RBV.
For patients treated for 12 wk, study visits took place at day 1 (baseline) and weeks 4, 8 and 12 (end-of-treatment, ΕOT). For patients treated for 24 w k, study visits took place at day 1 (baseline) and w eeks 4, 8, 12, 16, 20 and 24 (ΕOT). All patients had follow-up visits 4 and 12 w k after their last d ose of study med ication. Patients were consid ered to be off study if they d iscontinued early or did not comp lete p rotocol defined visits. The stud y protocol w as app roved by the Integ Review Institutional Review Board. The study was registered on ClinicalTrials.gov.
The primary endpoint was SVR 12 w k after the last treatment dose for the all treated population. The all treated population was defined as all patients that were consented and received at least 1 dose of stud y med ication. SVR w as d efined as HCV RNA below the low er limit of quantification (LLOQ) 12 w k after the end of treatment(SVR12). Plasma HCV RNA levels w ere measured using the COBAS TaqMan HCV test (version 2.0), for use with the High Pure System, which has an LLOQ of 25 IU per milliliter.
Outcomes for patients not achieving an SVR12 w ere record ed as on-treatment virologic failure (VF), post-treatment virologic relapse through post-treatment week 12 or failure d ue to other non-virologic reasons (e.g., premature d iscontinuation,adverse event, lost to follow-up, consent withdrawn). HCV RNA w as assessed at each stu d y v isit and p ost treatment w eek s 4 and 12 (or early p ost treatment discontinuation).
Key safety p arameters that w ere record ed includ ed d ose d iscontinuations/modifications due to adverse events (AΕs), treatment related serious AΕs (SAΕs), and laboratory test abnormalities. The onset and end dates, severity and relationship to study drug were recorded for each AΕ. Patients w ere questioned and/or examined by the investigator or a qualified d esignee for evidence of AΕs at each treatment visit.The determination of AΕ severity rested on medical judgment and w as made with the ap propriate involvement of the investigator or a d esignated sub-investigator. The severity of AΕs, with the exception of laboratory values, w as graded according to the WHO grad ing system. Investigators relied on clinical judgment in assigning severity to abnormal laboratory AΕs. Data on all treatment emergent AΕs were collected from the start of stud y d rug until 30 d after receipt of the last d ose. Clinical laboratory testing was performed at all visits during the treatment period.
The first pre-defined secondary endpoint evaluated the effect of baseline RASs on SVR12. Baseline samp ling for RASs w as obtained on d ay 1 for all p atients and thereafter only in patients w ith detectable HCV RNA after previously testing negative(breakthrough) or patients w ho w ere und etectable at the end of treatment but had detectable HCV RNA during the follow up period (relapse). Resistance testing w as performed by Quest Diagnostics. The patient subgroups evaluated included all RASs and different classes of RASs.
Another p re-d efined second ary end point evaluated p atient rep orted outcomes,examined via the Short Form-36 version 2 Health Survey (SF36v2), at baseline comp ared to end of treatment. Patients comp leted this self-ad ministered questionnaire, which assessed functional health and w ell-being at baseline and at 12 wk. The results consisted of eight scaled scores, which are the w eighted sums of the questions in their section. Scores w ere aggregated into a mental component summary(MCS) and a p hysical component summary (PCS); higher scores w ere indicative of better health.
The final secondary endpoint was to evaluate ad herence in patients receiving this treatment regimen. Pills w ere counted by stud y personnel at each treatment visit.Treatment compliance was defined as the subject having a total missed pill count of ≤20% of the total dispensed pill count over the course of their treatment duration. For patients on RBV, the total dispensed pill count w as 840 over 12 wk and 1680 over 24 wk. For patients not on RBV, the total dispensed pill count w as 336 over 12 wk and 672 over 24 wk. Patient adherence was reported according to treatment arm.
All patients w ho consented and received at least one d ose of study medication w ere included in the primary analysis for both efficacy and safety (all-treated population).Descrip tive summaries consisted of frequencies and p ercentages for categorical measures and of the number of p atients, mean, stand ard d eviation, med ian,minimum, and maximum values for continuous measures. Descriptive summaries were presented for select subgroups. Tabular summaries presented included age, sex,and race and other parameters measured at baseline. Since this w as a single arm study design, no power statement was calculated.
A total of 104 patients were screened and 100 patients were treated with the study drug regimens. The majority of patients (n = 89, 89%) were treated with OBV/PTV/r+ DSV + RBV, with 75 (75%) undergoing 12 wk of treatment and 25 (25%) undergoing 24 wk of treatment. The vast majority of patients (86%) w ere infected with GT1a(Figure 1). Patient characteristics at baseline, including current comorbidities and history of previous HCV treatments are shown in Table 1.
The final date for collection of data was October 3, 2017. One hundred (100) patients received at least one dose of study drug, 96 patients completed follow-up and 95(95%) achieved SVR12. The HCV RNA results for patients completing treatment with SVR are depicted in Figure 2. The vast majority (88.4%) were undetectable by week 4.Of the 4 patients who exited the study early, 3 were lost to follow-up and 1 was terminated early due to noncompliance. The one patient who completed treatment but did not achieve SVR was a 48-year-old Black male infected with GT1a HCV. He was treatment-naïve, FibroScan®showed no fibrosis (F0) and no baseline RASs were detected. His baseline viral load was 2694796 IU/mL and he was prescribed 12 wk of OBV/PTV/r + DSV + RBV. He had undetectable HCV RNA at treatment week 4 and treatment week 8; however, he missed 6 doses of RBV 8 wk into treatment. He was counseled on medication compliance and didn't miss any other doses. At end of treatment (week 12), HCV RNA was detected (742 IU/mL) and he was considered a treatment-failure. The patient did not return for his 12-wk follow-up visit.
The number of AΕs reported totaled 123. The most common AΕs were fatigue (12%),headache (10%), insomnia (9%) and diarrhea (8%) (Table 2). There were no serious AΕs or AΕs leading to discontinuation reported in this study.
Samp ling for RASs w as p erformed on 99 p atients at baseline. There w ere nodetectable mutations in the NS5a or NS5b genes in 88 (88%) and 98 (98%) of these patients, respectively. A probable resistance to OBV w as found in one patient (1%)w ith an NS5a mutation, but no resistance to d aclatasvir, elbasvir, led ip asvir or velpatasvir was noted. In add ition, there w ere 4 (4%) cases of possible resistance to OBV, daclatasvir, elbasvir, ledipasvir and velpatasvir. With regard to NS5b RASs, one patient (1%) had a possible resistance to DSV.
Table 1 Demographic and clinical characteristics of patients n (%)
Figure 1 Hepatitis C genotype and presence/absence of cirrhosis for patients. GT1: Genotype 1.
Patient reported outcomes were analyzed in patients with both baseline and end of treatment (Week 12 or Week 24) scores (n = 67). These results are portrayed in Table 3.Overall, PCS scores and MCS scores w ere significantly higher follow ing treatment compared to baseline (P = 0.04 and P = 0.011, respectively). Of the 8 scaled scores, all end of treatment scores w ere higher comp ared to baseline, w ith statistically significant improvement observed for 5 sub-scores (physical, general health, vitality,emotional and mental health).
Treatment compliance was assessed in all 100 patients and 98 (98%; 95% confidence interval, 93 to 99.4) of these p atients w ere consid ered comp liant. The tw o noncompliant patients were in the OBV/PTV/r + DSV + RBV 12-wk treatment arm.All patients in the OBV/PTV/r + DSV 12-wk arm and the OBV/PTV/r + DSV + RBV 24-wk treatment arm achieved 100% compliance. Four patients terminated the study early (prior to 12 wk) and are included in these data; all 4 met the criteria for treatment compliance while on treatment.
In a real w orld, all-comers p opulation of HCV GT1 p atients, a 12 or 24-w k multitargeted DAA regimen of OBV/PTV/r + DSV +/- RBV was highly effective w ith a 95% SVR12 rate. Of the 96 patients w ho completed follow up, 99% (95/96) achieved SVR12. The presence of baseline RASs had no impact on the ability to achieve SVR12.Furthermore, treatment w as associated w ith a low rate of treatment discontinuation unrelated to AΕs. No AΕs w ere considered serious. These efficacy and safety results are comparable to those seen in controlled clinical trials of OBV/PTV/r + DSV +/-RBV[8-11]. Although new er DAA regimens with shorter d urations have been recently app roved, our d ata on this old er regimen remains imp ortant; this regimen is an approved treatment option and is still used in some developing countries.
Our study population w as mad e up of slightly more males than females, w hich is consistent with the distribution of HCV in the United States[1]. Although there were no Asian patients and a small p ercentage of Black patients, this stud y rep resented an even distribution of Hispanics and non-Hispanics with HCV. HCV data are important in this population. Hisp anics are a large and fast-grow ing minority group in the United States[12]and, according to one study[13], Hispanics infected w ith HCV are at a significantly higher risk of developing cirrhosis and HCC than non-Hispanic Whites.
Figure 2 Hepatitis C virus RNA results for patients completing treatment with sustained virologic response.
Traditionally, individuals with HCV who have comorbidities are considered “hardto-cure” and are often exclud ed from controlled clinical trials. The HΕARTLAND stud y sought to enroll these ind ivid uals; many of the ind ivid uals stud ied had hyp ertension (44.2%), obesity (33.7%), Child-Pugh A cirrhosis (20.2%), steatosis(17.3%), type II diabetes mellitus (16.3%) and hyperlipidemia (16.3%). Including these p op ulations in HCV treatment stud ies is essential. Data has d emonstrated that,among patients w ith HCV, > 99% have at least one comorbid ity, w ith hypertension being one of the most common[14]. In addition, NHANΕs investigators predict that the incidence of cirrhosis w ill peak by the year 2020, affecting 1.76 million HCV-infected ind ivid uals in the United States[1]. Finally, metabolic conditions such as steatosis,obesity and diabetes are emerging as independent co-factors of fibrogenesis[15].
Ind ividuals who previously failed HCV treatment are also often disqualified from registration stud ies or evaluated in separate retreatment studies. Sixteen percent of patients in HΕARTLAND were HCV treatment experienced, w ith the majority (62.5%)previously treated w ith PegIFN + RBV. The inclusion of these patients did not affect overall SVR rates. This is consistent w ith w hat other studies of OBV/PTV/r + DSV+/- RBV have demonstrated: combining antiviral drugs w ith multiple mechanisms of action is effective regardless of the prior response to PegIFN + RBV[9,16].
In addition to including underrepresented HCV patients, data were collected in the“real w orld” setting, further d emonstrating its app licability in clinical p ractice.Moreover, results are similar to other published real-w orld data on this regimen. For example, in a study of 5168 HCV GT1 p atients, w ho w ere includ ed irrespective of cirrhosis status or prior HCV treatment exp erience, real w orld SVR12 rates w ere consistently high (94 to 99%) for the OBV/PTV/r + DSV +/- RBV regimen[17]. Another study p erformed in Taiw an show ed SVR12 rates of 98% and good tolerability in patients administered this regimen[18].
The data generated from the secondary analyses performed in this study are also consistent w ith published reports. With regard to resistance testing, according to the American Association for the Stud y of Liver Diseases (AASLD) HCV Guid ance,Recommend ations for Testing, Managing, and Treating Hepatitis C, a subset of patients w ith HCV will have viral variants harboring substitutions associated w ith resistance to DAAs, especially w ith NS5A inhibitor-containing regimens. This may negatively impact treatment response[19]. In HΕARTLAND, SVR rates app roached 100% d espite the presence of baseline NS5A and NS5B RASs in 12% and 2% of patients, respectively. This reinforces what is stated in the AASLD HCV guidance: the magnitude of the negative impact of RASs varies according to many factors and RAS testing alone will not dictate optimal DAA regimen selection[19].
Another second ary analysis w as the effect of OBV/PTV/r + DSV +/- RBV on patient reported outcomes. Such outcome assessments provid e patients' perspective on the impact of treatment on daily life and w ork. HΕARTLAND d emonstrated, via d ata collected from the SF36v2 short form, that OBV/PTV/r + DSV +/- RBV significantly imp roved p atient reported outcomes for total p hysical and mental components. The MALACHITΕ-I and MALACHITΕ-II controlled clinical studies in HCV G1 patients evaluated the same second ary end points using a similar methodand demonstrated matching trends. Overall, patients treated w ith OBV/PTV/r + DSV+/- RBV have better mental and physical health follow ing HCV treatment[20].
Table 2 Adverse events reported during the HEARTLAND study
Finally, p atient ad herence w as ad d ressed in HΕARTLAND. Ad herence is important w hen successfully treating HCV and becomes even more critical in the realw orld setting. Often, efficacy rates rep orted from clinical trials are substantially reduced when drugs are approved and used in clinical practice. The reasons for this are multifactorial and can be d ue to sid e effects, complexity of the regimen and/or other patient-related factors[21]. In our stud y, w e demonstrated excellent ad herence rates (98% treatment compliance) in the real-w orld setting using a complex regimen.This stud y was d esigned to enroll approximately 100 patients w ith baseline factors that may have limited their ability to enroll in registration trials including 28% with cirrhosis. The safety and efficacy results reported in this stud y are impressive and comparable to those reported in registration trials.
In conclusion, the all oral, DAA regimen containing OBV/PTV/r + DSV +/- RBV w as associated w ith a 99% SVR at p ost-treatment w eek 12 in GT1-infected patients,with and without compensated cirrhosis.
Table 3 Mean short form 36 version 2 scores using the normative based scores
The hepatitis C virus (HCV) is a prevalent virus that, if left untreated, leads to chronic liver disease and, ultimately, death. The new era of direct acting antiviral (DAA) treatment regimens has the potential to cure the virus [i.e., achieve sustained virologic response (SVR)] in the majority of patients. The HCV NS5A inhibitor ombitasvir (OBV), HCV NS4/4A protease inhibitor p aritaprevir (PTV), the CYP3A inhibitor ritonavir and the non-nucleoside NS5B polymerase inhibitor dasabuvir (DSV) (OBV/PTV/r + DSV) w ith or w ithout ribavirin (RBV) is a DAA regimen that achieves SVR rates as high as 99% in HCV genotype 1 (GT1) patients in controlled clinical studies. How ever, there are patients w ho are considered “hard to cure” that are trad itionally excluded from registration trials due to rigorous study inclusion criteria,presence of comorbidities and previous treatment failures. This phase 4, open label study evaluated the safety and efficacy of OBV/PTV/r + DSV +/- RBV in a real-w orld clinical setting in patients w ho have historically been excluded from clinical trials. This study is completed.
Controlled clinical studies demonstrate 99% SVR rates in patients w ith HCV GT1, how ever,many patients in these studies do not meet the inclusion criteria for these studies. In a real world population of HCV patients, many have comorbidities or history of previous HCV treatment failures. We sought to examine the efficacy and safety of OBV/PTV/r + DSV +/- RBV in real w orld HCV patients w ho are generally und errepresented in clinical trials. This study also examined patient quality of life, d osing ad herence and w hether resistance-associated substitutions (RASs) impact achievement of SVR, w hich are all real w orld issues encountered in HCV patients. The results of this study w ill determine if controlled clinical trial results can be expected in everyday HCV patients seen in clinical practice.
The primary objective of this study was to examine the efficacy and safety of OBV/PTV/r + DSV+/- RBV in real world HCV patients generally underrepresented in clinical trials. This study found that this treatment regimen was highly effective and no adverse events were considered serious; these results are comparable to those seen in controlled clinical trials with this treatment regimen. Therefore, including patients w ith comorbidities or a history of previous HCV treatment(s) did not affect the results. According to this one study, the results demonstrated in controlled clinical trials involving OBV/PTV/r + DSV +/- RBV can be applied to everyday HCV patients seen in clinical practice.
Patients w ere ≥ 18 years old and chronically infected w ith HCV GT1 (GT1a, GT1b or GT1a/1b).Patients w ere treatment-naïve or p rev iously failed a regimen includ ing p egylated interferon/RBV +/- telaprevir, boceprevir, or simeprevir. One hundred patients were treated w ith the study drug regimen, which w as administered for 12 or 24 w k +/- RBV according to GT1 subtype and presence/absence of cirrhosis. Patients w ere evaluated every 4 w k from treatment day 1 and at 4 and 12 wk after end-of-treatment.
Many of the patients stud ied had comorbidities (44.2% hypertensive, 33.7% obese, 20.2%cirrhotic) and 16% previously failed HCV treatment. Ninety-six patients completed study followup and 99% achieved 12-w k sustained virologic response. The majority (88.4%) of patients had und etectable HCV RNA by w eek 4. The most common adverse events w ere fatigue (12%),headache (10%), insomnia (9%) and diarrhea (8%); none led to treatment discontinuation.Physical and mental patient reported outcomes scores significantly improved after treatment.Almost all (98%) patients w ere treatment compliant.
In an all-comers HCV GT1 population, 12 or 24-w k of OBV/PTV/r + DSV +/- RBV is highly effective and tolerable and results in better mental and physical health following treatment.
Results of he approved use of the OBV/PTV/r + DSV +/- RBV regimen in HCV GT1 patients in clinical practice can potential mirror results obtained in controlled clinical trials. The availability of real w orld data on approved HCV treatment regimens is extremely useful in clinical practice.New er DAA regimens w ith shorter treatment durations have been recently approved. These regimens should also be evaluated in the real w orld population of HCV patients. Future clinical studies need to evaluate the efficacy and safety of these new er DAA regimens in real w orld patients. Patients with comorbidities and those who have had previous HCV treatment failures should be included in these studies. In addition, secondary measures should include physical and mental outcomes, the affects of RASs and adherence to the newer regimens.
Data management w as performed by John Sendajo of Texas Liver Institute. Statistical analyses w ere performed by Marc Schwartz of MS Biostatistics, LLC. Writing support was provided by Rachel Ε. Bejarano, Pharm D of RB Scientific Services, LLC and Lisa D. Pedicone, PhD of Cerulean Consulting, LLC.
World Journal of Gastroenterology2019年18期