Efficacy of vonoprazan against bleeding from endoscopic submucosal dissection-induced gastric ulcers under antithrombotic medication: A cross-design synthesis of randomized and observational studies.

Vonoprazan, a potassium-competitive acid blocker, is expected to be superior to proton pump inhibitors (PPIs) in preventing post-endoscopic submucosal dissection (ESD)-induced gastric bleeding. However, the results of randomized controlled trials (RCTs) and observational studies on the efficacy of vonoprazan have been inconsistent. This study aimed to evaluate the effectiveness of vonoprazan in antithrombotic drug users, a population that has been excluded from RCTs. Treatment effects were assessed using cross-design synthesis, which can be adjusted for differences in study design and patient characteristics. We used data from an RCT in Japan (70 patients in the vonoprazan group and 69 in the PPI group) and an observational study (408 patients in the vonoprazan group and 870 in the PPI group). After matching, among the antithrombotic drug users in the observational study, post-ESD bleeding was noted in 8 out of 86 patients in the vonoprazan group and 18 out of 86 patients in the PPI group. After pooling the data from the RCT and observational study, the risk difference for antithrombotic drug users was -14.6% (95% CI: -22.0 to -7.2). CDS analysis suggested that vonoprazan is more effective than PPIs in preventing post-ESD bleeding among patients administered antithrombotic medications.

Introduction

Endoscopic submucosal dissection (ESD) of gastric neoplasms is a minimally invasive and well-established procedure [1]. However, post-ESD bleeding from artificial ulcers remains an unsolved problem [2-4]. Vonoprazan (Takeda Pharmaceutical Co., Ltd., Tokyo, Japan) directly inhibits proton pumps without requiring activation by gastric acid. In addition, it rapidly and strongly suppresses acid secretion [5]. Vonoprazan has a dissociation coefficient of 9.37 and is more likely to accumulate in the secretory tubules of the stomach than proton pump inhibitors (PPIs) [6]. In addition, vonoprazan offers several advantages over PPIs that allow it to achieve a higher gastric pH. These include rapid drug absorption after administration (median Tmax, 2 h), longer plasma half-life than PPIs, and no interindividual pharmacokinetic variations due to differences in cytochrome P450 2C19 (CYP2C19) genotypes [7]. A crossover study demonstrated that vonoprazan could suppress gastric acid secretion rapidly and more strongly than PPIs [8]. Therefore, vonoprazan is expected to be more effective than PPIs for treating artificial gastric ulcers resulting from ESD. Our previous observational study showed that vonoprazan suppressed post-ESD bleeding more effectively than PPIs [9]. However, subsequent randomized controlled trials (RCTs) comparing PPIs and vonoprazan in terms of post-ESD bleeding control did not find vonoprazan to be superior to PPIs [10-15].

With an increase in the aging population, the number of patients being administered antithrombotic drugs is increasing in developed countries. The proportion of patients undergoing gastric ESD who receive antithrombotic drugs is also expected to increase [16]. A meta-analysis revealed that antithrombotic drug users have a significantly increased risk of post-ESD bleeding than non-users, regardless of whether they suspended the treatment perioperatively. However, the analysis only included studies in which PPIs were used to treat post-ESD ulcers [17]. To improve the safety of gastric ESD, it is important to determine whether vonoprazan suppresses post-ESD bleeding more effectively than PPIs in individuals consuming antithrombotic drugs.

A subgroup analysis in our observational study suggested that vonoprazan suppresses post-ESD bleeding in antithrombotic drug users [9]. Safety-critical RCTs generally exclude patients at high risk of trial-related complications. Indeed, RCTs with post-ESD bleeding as the primary endpoint excluded antithrombotic drug users [10, 11]. In RCTs that included antithrombotic drug users, it was difficult to draw firm conclusions regarding the prevention of post-ESD bleeding for several reasons. These reasons include the primary endpoint being ulcer reduction, small sample size, inclusion sample size of antithrombotic drug users, low post-bleeding rates, and slow initiation of vonoprazan [12-15].

In real-world clinical practice, complications associated with ESD have major consequences in elderly individuals and patients with serious comorbidities who cannot be included in RCTs. Cross-design synthesis (CDS) is a method for estimating the treatment effect in patients excluded from RCTs. It involves pooling the data of an observational study and an RCT to estimate the treatment effect in the overall treated population [18, 19]. In the present study, we used CDS to evaluate the inhibitory effect of vonoprazan on post-ESD gastric ulcer bleeding in patients consuming antithrombotic drugs.

Methods

Study design and data collection

In this study, we used CDS, a method that pools the results of an observational study and an RCT, to estimate treatment effects in patients excluded from the RCT and the overall treated population. Specifically, the observational study included subgroups that met the exclusion criteria of the RCT. The CDS method, designed to address the generalizability problem in a real-world clinical setting caused by sample selection bias in RCTs, is used to estimate the effect of a treatment on a patient population that is unable to participate in RCTs using observational data. As the patients in the observational study were not randomized, the results could be biased due to treatment selection errors. However, the CDS estimator is unbiased if the assumption that the treatment selection error for stratified estimators from the observational study is constant across strata [18, 19]. The details of the assumptions that the CDS estimator is unbiased are shown in the S1 and S2 Appendices. This study evaluated the effectiveness of vonoprazan in patients taking antithrombotic drugs, focusing on the exclusion criteria for antithrombotic drug use. Antithrombotic drugs were defined as taking either antiplatelet or anticoagulant drugs, and data based on the study by Kagawa et al. that included patients taking antithrombotic drugs were used for the observational study. A summary of the pooled studies is presented in Table 1. This study was approved by the Ethics Committee of the Graduate School and Faculty of Medicine, Kyoto University (approval number: R2141) and the Ethics Committee of Kagawa Prefectural Central Hospital (approval number: 839), and adhered to the Declaration of Helsinki and Ethical Guidelines for Medical and Health Research Involving Human Subjects.

Table 1

Summary of pooled studies.

	RCT (Hamada et al., 2019)	Observational study
Study design	Single-centre, randomized phase II trial	Retrospective observational cohort study
Study period	2015–2016	2005–2018
Research facilities	Osaka International Cancer Institute	Kagawa Prefectural Central Hospital
Inclusion criteria	(i) Age 20 years or older(ii) ECOG performance status 0–2(iii) Hb≥9 g/dl; Plt≥100,000/ mm 3; AST,ALT≤100 U/l; Cre≤2.0 mg/dl; PT≥70%.	Age 20 years or older
Exclusion criteria	(i) Concurrent endoscopic treatment for esophageal or duodenal lesion(ii) History of endoscopic treatment in the past 28 days(iii) Endoscopic treatment scheduled within 28 days after gastric ESD(iv) History of gastric resection(v) allergy to vonoprazan or lansoprazole(vi) Systemic administration of corticosteroids, anticoagulant agents, or antiplatelet agents(vii) Systemic administration of nonsteroidal anti-inflammatory drugs that could not be suspended between 7 days before gastric ESD and 28 days afterwards(vii) Pregnancy or lactation(ix) Major organ failure	(i) Remnant stomach(ii) Surgical resection of the stomach or chemotherapy after ESD(iii) Discontinuation of anti-ulcer agents for any reason after ESD
Endoscopist performing ESD	Either an experienced endoscopist or a resident under the supervision of an experienced endoscopist	One of two board-certified endoscopists who had previously performed ESDs in over 100 gastric neoplasm cases
Instrument used to dissect gastric neoplasms	The insulated-tipped knife-2 (Olympus Medical Systems, Co. Ltd., Tokyo, Japan) or a Flush Knife(Fuji Film Medical, Tokyo, Japan)	The insulated-tipped knife-2 (Olympus Optical Co., Tokyo, Japan)
Coagulation of ulcers	VIO 300D, ERBE Elektromedizin, Tübingen, Germany	Haemostatic forceps (Coagrasper; Olympus Optical Co.)

RCT, Randomised controlled trial; ESD, Endoscopic submucosal dissection; VPZ, Vonoprazan; PPI, Proton pump inhibitor; ECOG, Eastern Cooperative Oncology Group; Hb, hemoglobin; Plt, platelet count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; Cre, Creatinine; PT, prothrombin time.

Randomized clinical trial

To identify studies eligible for analysis, a literature search was conducted using the PubMed, Scopus, and Cochrane library databases for all articles published through November 2020. The initial search terms included ‘vonoprazan’, ‘Takecab’, ‘TAK-438’, ‘potassium-competitive inhibitor’, ‘ESD’, and ‘endoscopic submucosal dissection’. An example of a full electronic search strategy used for the online database is shown in S1 Table and a flow diagram for assessing the studies is shown in Fig 1.

Fig 1

Flow diagram of assessment of randomized controlled trials used in CDS.

RCTs, Randomised controlled trials; ESD, Endoscopic submucosal dissection.

A total of 101 articles were identified by literature search, and after letters, meta-analyses, and duplicate articles and protocols were excluded, six RCTs were eligible for CSD pooling. All six RCTs compared the efficacy of vonoprazan with PPIs for post-ESD bleeding. Vonoprazan was administered 1 day before ESD in three studies, and a few days after ESD in the other three. The former group of RCTs was considered eligible for CDS because these studies used the same vonoprazan dosing regimen as the study by Kagawa et al. [9] (i.e., vonoprazan administration the day before ESD). We decided to use the RCT by Hamada et al., which set the primary endpoint as bleeding after ESD and calculated the planned sample size [10].

The study by Hamada et al. was an open-label, prospective, randomized phase II clinical trial conducted at the Osaka International Cancer Center in Japan [10]. One hundred and forty patients who underwent ESD between 2015 and 2016 were enrolled, and patients taking antithrombotic drugs were excluded. Patients were randomized in a 1:1 ratio to vonoprazan (n = 69) or PPI (n = 70).

Observational study

For the observational study data, we used the data of ESD cases performed at our hospital from 2005 to 2018. The study by Kagawa et. al. used data up to 2015, but the sample size of 17 patients taking antithrombotic drugs was small and considered insufficient to examine the effect in patients taking antithrombotic drugs. Therefore, in this study, we used data up to 2018, when the sample size of antithrombotic and non-antithrombotic patients was more than 70 cases per group, similar to the sample size design of Hamada et al.

The observational study data analyzed in the CDS included patients taking antithrombotic drugs. In this study, 408 patients who underwent ESD and were treated with vonoprazan for postoperative ulcer treatment at Kagawa Prefectural Central Hospital in Japan from April 2014, when vonoprazan was launched in April 2018 were prospectively enrolled in the vonoprazan group. Then, from among 870 patients who underwent ESD and were treated with PPIs from 2005 to 2014, before the launch of vonoprazan, were analyzed as historical controls. All participants provided written informed consent, and they were given the opportunity to opt out of the present study.

Treatment and follow-up

When performing CDS, the outcomes and study protocols of the studies to be pooled should be largely consistent. The treatment schedules in the RCT and observational studies are shown in Fig 2. The withdrawal schedules for each antithrombotic drug in the observational study are shown in Fig 3.

Fig 2

Treatment schedules of the randomized clinical trial and observational study.

RCT, Randomised controlled trial; ESD, Endoscopic submucosal dissection; VPZ, Vonoprazan; PPI, Proton pump inhibitor; iv, Intravenous.

Fig 3

The withdrawal schedules of each antithrombotic drug in the observational study.

ESD, Endoscopic submucosal dissection.

Participants in both studies were admitted the day before ESD, and all participants were hospitalized for 7 days. In both studies, the vonoprazan group started oral vonoprazan 20 mg on the day before ESD [9, 10]. The PPI group in the RCT started oral lansoprazole 30 mg the day before ESD, but those in the observational study received omeprazole intravenously for 2 days (day 0 and day 1, 40 mg/day) and then started oral rabeprazole (from day 2, 20 mg/day) [9, 10]. An infusion of omeprazole is known to rapidly raise the pH in the stomach to above 4.2 within a few hours [20]. On the other hand, oral PPIs require several days of dosing to show effects because they must be activated by acid [21, 22]. Therefore, the PPI groups in the two studies can be regarded as roughly equivalent regarding the PPI doses intended to produce an effect on the day of ESD.

Vonoprazan was administered for 8 weeks in the RCT and 5 weeks in the observational study. Most post-ESD bleeding events develop within the first two weeks after ESD [23]. In the study by Hamada et al., the Kaplan-Meier method was used to determine the cumulative non-bleeding rate within 28 days after ESD, and all bleeding occurred within 14 days after ESD [10]. Data from the observational study were used to analyze the outcomes up to 28 days after ESD. Thus, the durations of drug administration and outcome observations in the two studies were the same.

Study outcome

The primary outcome of this study was post-ESD bleeding, which was defined as hematemesis, melena, or a decrease in hemoglobin of ≤ 2 g/dl in both the RCT and observational studies [9, 10].

Statistical analysis

The CDS in this study used the estimate of the treatment effect in patients not taking antithrombotic drugs, derived from the RCT, and the estimate of the treatment effect stratified by use of antithrombotic drugs, derived from the observational study. For the estimate from the RCT, we used the risk difference shown in Hamada et al. [10].

Propensity score matching (PSM) was performed to avoid the effects of measurable confounding factors when using observational study data. One-to-one matching without replacement was completed using the nearest neighbor match on the logit of the propensity score for the treatment approach with a caliper width set to 0.2 times the standard deviation of the logit of the propensity score. To estimate the propensity score, we fitted a logistic regression model for treatment as a function of patient characteristics, specifically age, Helicobacter pylori infection, tumor size, and tumor location. PSM was performed in the overall treated patients, patients taking antithrombotic drugs, and patients not taking antithrombotic drugs, and the risk difference in the matched data was calculated for each patient population.

Summary statistics were constructed using frequencies and proportions for categorical data and means and standard deviations for continuous data. Standardized differences were estimated for the factors used in the propensity score model after matching to assess post-match balance. Standardized differences of less than 10.0% for a given covariate indicate a relatively small imbalance [24]. The Wald confidence interval (CI) was used to indicate risk differences in the observational study. Confidence intervals for the CDS-based estimates were calculated using the method presented by Kaizer [18]. All statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA) and R version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria).

Results

Study population

We analyzed the RCT data from Hamada et al. comprising [10] 70 patients in the vonoprazan group and 69 in the PPI group along with the observational study data of 408 patients in the vonoprazan group and 870 in the PPI group. The patient characteristics from each study are shown in Table 2. With the use of PSM in the overall treated patients, 392 patients in the vonoprazan group were matched with 392 patients in the PPI group. Similarly, among the patients who were administered antithrombotic drugs, 86 patients in the vonoprazan group were matched with 86 patients in the PPI group. In addition, among patients not consuming antithrombotic drugs, 306 patients in the vonoprazan group were matched with 306 patients in the PPI group. Out of the total number of treated patients, 16 patients (nine patients taking antithrombotic medications and seven patients not taking antithrombotic medications) were dropped because of caliper matching. Among those that were matched, the standardized differences were below 10.0% for all factors, indicating only small differences between the two groups (Table 3).

Table 2

Patient characteristics in the randomized clinical trial and observational study before propensity score matching.

Characteristic	RCT (Hamada et al., 2018)		Observational study
	VPZ, n (%)	PPI, n (%)	VPZ, n (%)	PPI, n (%)
Total	69	70	408	870
Age, mean ± SD	70.3 ± 6.8	70.1 ± 8.2	73.1 ± 8.9	71.4 ± 9.1
Male	51 (74%)	57 (81%)	295 (72.3%)	646 (74.3%)
Positive for Helicobacter pylori	27 (39%)	29 (41%)	142 (34.8%)	366 (42.1%)
Endoscopic tumor size (mm), mean ± SD	17 ± 13	16 ± 10	36.7 ± 13.4	35.4 ± 12.9
Tumor size >2 cm	12 (17%)	11 (16%)	376 (92.2%)	813 (93.4%)
Tumor located in the antrum	34 (49%)	36 (51%)	260 (63.7%)	477 (54.8%)
Oral anti-thrombotic drug	-	-	95 (23.3%)	106 (12.2%)

RCT, Randomised controlled trial; VPZ, Vonoprazan; PPI, Proton pump inhibitor; SD, Standard deviation.

Table 3

Patient characteristics in the observational study after propensity score matching according to antithrombotic drug treatment.

Characteristic	Overall treated patients			Taking antithrombotic drugs			Not taking antithrombotic drugs
	VPZ, n(%)	PPI, n(%)	Standardized difference, %	VPZ, n(%)	PPI, n(%)	Standardized difference, %	VPZ, n(%)	PPI, n(%)	Sstandardized difference, %
Total	392	392	-	86	86	-	306	306	-
Age, mean ± SD	72.6 ± 8.7	72.3 ± 8.8	2.6	75.3 ± 8.9	75.0 ± 9.1	3.5	72.1 ± 8.9	71.9 ± 9.1	1.9
Positive for Helicobacter pylori	142 (36.2%)	142 (36.2%)	0.0	35 (40.7%)	38 (44.2%)	7.1	105 (34.3%)	108 (35.3%)	2.1
Tumor size >2 cm	368 (93.9%)	362(92.3%)	6.0	84 (97.7%)	84 (97.7%)	0.0	286 (93.5%)	285 (93.1%)	7.9
Tumor located in the antrum	244 (62.2%)	245 (62.5%)	0.5	56 (65.1%)	56 (65.1%)	0.0	188 (61.4%)	194 (63.4%)	4.0

VPZ, Vonoprazan; PPI, Proton pump inhibitor; SD, Standard deviation.

Primary outcome

The efficacy of vonoprazan compared with that of PPIs in preventing post-ESD bleeding is summarized in Table 4. Based on observational study data, the bleeding rate was 5.4% in the vonoprazan group and 5.6% in the PPI group. The estimated treatment effect among all the treated patients (the risk difference), including those taking and not taking antithrombotic drugs, was -0.3% (95% CI: -3.4% to −2.9%). In contrast, the estimate obtained by pooling data from the RCT and the observational study using CDS was -4.3% (95% CI: -11.7% to 3.1%). The bleeding rate in patients not taking antithrombotic medication was 4.3% in the RCT and 3.9% in the observational study for the vonoprazan group, 5.7% in the RCT, and 2.3% in the observational study for the PPI group. The estimated treatment effect was -1.4% (95% CI: -10.3% to 7.3%) in the RCT and 1.6% (95% CI: -1.1% to 4.4%) in the observational study. The bleeding rates among patients taking antithrombotics were 9.3% and 20.9% in the vonoprazan and PPI groups, respectively. The estimated treatment effect was -11.6% (95% CI: -22.2% to -1.1%) based on the observational study; however, the estimate obtained using CDS was -14.6% (95% CI: -22.0% to -7.2%).

Table 4

Efficacy of vonoprazan compared with proton pump inhibitors in preventing post-ESD bleeding according to antithrombotic drug treatment.

	Post-ESD bleeding		Risk difference (%)	95% confidence interval
	VPZ, n (%)	PPI, n (%)
Overall treated patients
Observational study	21/392 (5.4%)	22/392 (5.6%)	-0.3	-3.4 to 2.9
Cross-design synthesis	-		-4.3	-11.7 to 3.1
Patients not taking antithrombotic drugs
RCT	3/69 (4.3%)	4/70 (5.7%)	-1.4	-10.3 to 7.3
Observational study	12/306 (3.9%)	7/306 (2.3%)	1.6	-1.1 to 4.4
Patients taking antithrombotic drugs
Observational study	8/86 (9.3%)	18/86 (20.9%)	-11.6	-22.2 to -1.1
Cross-design synthesis	-		-14.6	-22.0 to -7.2

ESD, Endoscopic submucosal dissection; VPZ, Vonoprazan; PPI, Proton pump inhibitor; RCT, Randomised controlled trial

The results of post-ESD bleeding according to the type of antithrombotic drug used are shown in Table 5. Of the 86 patients in the vonoprazan group, 45 were on antiplatelet monotherapy, such as low-dose aspirin (LA)—32 were on LA, 11 on thienopridine, and 2 on other antiplatelet agents. Twenty-six patients received dual antiplatelet therapy (four received LA and thienopridine, seven received LA and other antiplatelet agents, and 15 received two other antiplatelet agents), of which two patients suffered post-ESD bleeding. Eight patients were on anticoagulant monotherapy (warfarin) and three were on direct oral anticoagulants (DOACs) (two on rivaroxaban and one on dabigatran), of which three and zero patients, respectively, experienced post-ESD bleeding. Three patients were on both, an antiplatelet agent and an anticoagulant (warfarin) (one on LA and warfarin, two on other antiplatelet agents and warfarin) and two patients were on both, an antiplatelet agent and an anticoagulant (DOAC) (one on other antiplatelet agents and apixaban, one on LA and dabigatran), of which zero and one patient, respectively, showed post-ESD bleeding. Of the 86 patients in the PPI group, 37 were on antiplatelet monotherapy (23 on LA, 10 on thienopridine, 4 on other antiplatelet agents) and 24 patients were on dual antiplatelet therapy (4 on LA and thienopridine, 7 on LA and other antiplatelet agents, 12 on two other antiplatelet agents), of which 3 and 7 patients, respectively, experienced post-ESD bleeding. Seventeen patients were on anticoagulant monotherapy (warfarin), of which three patients suffered post-ESD bleeding. Seven patients were on both, an antiplatelet agent and an anticoagulant (warfarin) (four on LA and warfarin, three on other antiplatelet agents and warfarin) and one patient was on both, an antiplatelet agent and an anticoagulant (DOAC) (LA and rivaroxaban), of which four and one patient, respectively, experienced post-ESD bleeding.

Table 5

Post-ESD bleeding by types of antithrombotic drugs taken.

	VPZ, n (%)	PPI, n (%)	Risk difference (%)	95% confidence interval
Total	8 / 86 (9.3%)	18 / 86 (20.9%)	-11.6	-22.2 to -1.1
Anti-platelet monotherapy	2 / 45 (4.4%)	3 / 37 (8.1%)	-3.7	-14.3 to 7.0
Dual anti-platelet agents	2 / 26 (7.7%)	7 / 24 (29.2%)	-21.5	-42.4 to -0.01
Anti-coagulant monotherapy (warfarin)	3 / 8 (37.5%)	3 / 17 (17.6%)	19.9	-18.3 to 57.9
Anti-coagulant monotherapy (DOAC)	0 / 3 (0.0%)	- / 0	-	-
Combination of anti-platelet and anti-coagulant (warfarin)	0 / 3 (0.0%)	4 / 7 (57.1%)	-57.1	-93.8 to -20.5
Combination of anti-platelet and anti-coagulant (DOAC)	1 / 2 (50.0%)	1 / 1 (100.0%)	-50.0	-100.0 to 19.3

ESD, Endoscopic submucosal dissection; VPZ, Vonoprazan; PPI, Proton pump inhibitor; DOAC, Direct oral anticoagulant

Discussion

An observational study by Kagawa et al. [9] suggested that vonoprazan significantly reduced post-ESD bleeding compared to PPIs. However, these results were inconsistent with the results of the RCT conducted by Hamada et al. [10], which led us to perform the present study.

In their observational study, Kagawa et al. revealed that the risk difference of post-ESD bleeding in patients consuming antithrombotic drugs between vonoprazan and PPIs was -36.7%; however, the sample size for patients taking antithrombotic drugs was insufficient. In addition, the robustness of evidence from observational studies is limited. Therefore, in the present study, we increased the sample size of patients receiving antithrombotic drugs and examined the results using the CDS method. This guided treatment selection bias in the observational study data. Subsequently, the risk difference in patients taking antithrombotics was -11.6%, while that using the CDS method was -14.6%. Regarding tumor size, the patients enrolled in the observational study had a larger tumor size than those enrolled in the RCT. Therefore, the estimates of treatment effects from observational study data may be biased due to higher bleeding rates. However, since the percentage of patients with large tumors was similar between patients with and without antithrombotic medication in the observational study, the CDS estimator may have excluded such bias, by the definition of sufficient conditions under which the CDS estimator is unbiased.

The two main types of anticoagulants are warfarin and DOACs. In two studies consisting of a small number of DOAC users, the frequency of post-ESD bleeding events was reported to be 20.8% [25] and 22.0% [26], respectively. The risk of post-ESD bleeding was reportedly higher with warfarin than with DOACs in high-risk endoscopic procedures. However, there was no difference in the risk of bleeding during upper gastrointestinal ESD [27]. In addition, the risk of post-ESD bleeding differs between individual DOAC users [25]. Since only six DOAC users were included in this study, further studies are needed to determine the risk of post-ESD bleeding with specific anticoagulant drugs.

In this study, most of the enrolled antithrombotic drug users were receiving a single antiplatelet agent. A previous meta-analysis revealed that single antiplatelet agent use was also a risk factor for post-ESD bleeding, with an odds ratio of 2.08 compared to non-users [17]. Other studies have shown that the odds ratios for post-ESD bleeding in patients who received double antiplatelet therapy [28], warfarin [29], or LDA and warfarin [30] were 5.06, 10.15, and 14.38, respectively, when compared with non-antithrombotic drug users. In this study, following PSM, we assessed the difference in the incidence of post-ESD bleeding between vonoprazan and PPIs according to the type of antithrombotic drug used. The post-ESD bleeding rate with the combined use of an antiplatelet and anticoagulant (warfarin or DOAC), which are presumed to pose the highest risk of bleeding, was lower in the vonoprazan group than in the PPI group (20.0% vs. 62.5%). The superiority of vonoprazan over PPI in preventing post-ESD bleeding may, therefore, be more pronounced in patients at a high risk of bleeding.

Several attempts have been made to prevent post-ESD bleeding. Long-term discontinuation of antithrombotic drugs before and after gastric ESD may help avoid increasing the risk of post-gastric ESD bleeding caused by antithrombotic drugs. Several retrospective studies have reported that discontinuing antithrombotic therapy a week before gastric ESD does not increase the risk of thromboembolism [31, 32]. However, other studies have shown that prolonged discontinuation can increase the risk of thromboembolism [33, 34]. Traditionally, bridging therapy from oral warfarin to intravenous heparin has been used during the perioperative period of ESD, considering both thromboembolism and post-ESD bleeding. Unfortunately, bridging therapy in anticoagulant users is not appropriate for gastric ESD because it does not decrease every complication [25, 27]. Second-look endoscopy (SLE) has been performed after ESD to evaluate the ulcer condition and to perform preventive hemostatic treatment for high-risk bleeding ulcers [35]. However, SLE cannot suppress bleeding after ESD, and bleeding prevention procedures performed during SLE may be harmful [36, 37]. Therefore, other methods are necessary to suppress post-ESD bleeding in antithrombotic drug users.

In upper gastrointestinal bleeding, the intragastric pH significantly affects the coagulation system (including platelets), which is involved in hemostasis. An intragastric pH of 5.4 or higher is required for blood coagulation to achieve hemostasis in peptic ulcers [38] because pepsin causes platelet destruction at pH 5.0–5.5 [39]. This suggests that maintaining an intragastric pH of at least 5.4 is extremely important for suppressing post-ESD bleeding. Treatment for ESD-induced artificial ulcers with gastric acid inhibitors is usually initiated on the date of ESD. However, post-ESD bleeding occurs most frequently within 24 hours after the procedure [5, 35]. In the observational study analyzed here, vonoprazan was administered 1 day before ESD to achieve a high intragastric pH on the day of performing the technique, thereby facilitating effective hemostatic function. The RCT analyzed in the present study also involved the administration of vonoprazan in the same way.

Concerning gastric acid suppression, PPI administration can also achieve an intragastric pH of 5 or more, even when administered intravenously or orally. In the observational study, a combination of 2-day intravenous infusions of omeprazole and an oral intake of 20 mg of rabeprazole was adopted in the PPI group. A single infusion of omeprazole is known to quickly increase the intragastric pH >4.25 [20]. On the other hand, for the PPI group in the RCT, administration of oral lansoprazole 30 mg was started the day before ESD was conducted. The ratios of post-ESD bleeding in non-antithrombotic drug users were low and almost the same in the observational study and RCT (2.3% vs. 5.7%, respectively), suggesting that a sufficiently high intragastric pH was achieved by PPI administration on the day of ESD, to provide adequate prophylaxis against bleeding in non-high-risk patients.

A crossover study compared vonoprazan with a PPI in terms of inhibiting gastric acid secretion. The holding time ratios at an intragastric pH of 5 or higher on day 1 after administration of vonoprazan and esomeprazole were 62.8% and 13.0%, respectively, and those for vonoprazan and rabeprazole were 76.6% and 16.7%, respectively. However, esomeprazole and rabeprazole were better able to achieve an intragastric pH of 5 or higher after 7 days of administration (48.3% and 53.2%, respectively) [8]. Although initiating PPI therapy 7 days before ESD is difficult in everyday clinical practice, it might further reduce the risk of bleeding in antithrombotic drug users.

Because of various types of biases, patients with medical backgrounds that might affect the results were excluded from the study. However, it has been reported that the results of RCTs obtained in this way differ from those observed in real-world clinical practice [40-42]. The CDS method was designed by the United States General Accounting Office to address the generalizability problem caused by sample selection bias in RCTs [43]. Recently, the statistical properties of the CDS estimator have been evaluated by Kaizer [18]. In the simulation studies, the CDS estimates had a lower bias than the commonly used estimates based on randomized or observational studies alone under reasonable data assumptions. Therefore, very few studies have employed CDS to analyze real-world clinical data [18, 44]. In this study, CDS analysis suggested that vonoprazan is more effective than PPIs in preventing post-ESD bleeding in patients taking antithrombotic drugs. It may be difficult to perform RCTs targeting antithrombotic drug users with various comorbidities. However, we believe that future observational studies will confirm the validity of the results obtained in this study.

This study had several limitations. First, even if the confounding effects could have been examined by CDS, selection bias cannot be ruled out because both RCT and observational studies were from a single center. CDS estimates may depend on the data used; therefore, it is necessary to examine a variety of data. In addition, when integrating data from two studies, all items were not the same in the design of the observational studies and RCTs. Second, the CDS method cannot account for all unknown confounders, and the validity of sufficient conditions cannot be confirmed from the data. Therefore, the bias of the estimates using CDS is not always zero, and further improvement of the method is desired. Third, although the same experienced endoscopists performed ESD in the PPI and vonoprazan groups in the observational study, the results may have been influenced by the endoscopists’ experience. Finally, neither of the two studies analyzed here investigated the relationship between gastric pH and post-ESD bleeding.

Example of full electronic search strategy used for the online database.

(PDF)

Click here for additional data file.

The CDS estimator.

(PDF)

Click here for additional data file.

Sufficient assumptions that the CDS estimator is unbiased.

(PDF)

Click here for additional data file.

1 Oct 2021

PONE-D-21-19373Efficacy of vonoprazan against bleeding from endoscopic submucosal dissection-induced gastric ulcers under antithrombotic medication: A cross-design synthesis of randomized and observational studiesPLOS ONE

Dear Dr. Hidaka,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Nov 15 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Sanjiv Mahadeva, MRCP, MD

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf.

2. Please ensure you have provided the full electronic search strategy for at least one database

3. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

4. Thank you for stating the following in the Acknowledgments/ Funding Section of your manuscript:

“This study was funded by AMED under Grant Numbers JP19lk0201061 and 340

JP20lk0201061.”

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

“The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

5. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

Additional Editor Comments (if provided):

The study is of clinical interest, but the manuscript needs some revision. Please respond to the reviewers' comments. The Discussion needs to be simplified as well.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dear Editors and authors,

I have read the manuscript PONE-D-21-19373 entitled “Efficacy of vonoprazan against bleeding from endoscopic submucosal dissectioninduced gastric ulcers under antithrombotic medication: A cross-design synthesis of randomized and observational studies” with interest. The authors used the uncommon but interesting statistic method in data analysis to evaluate the efficacy of vonoprazan in preventing post ESD bleeding in patients taking anti-platelet and/or anti-coagulant, compared to PPI. However, there are many important major concerns that need to clarify in the manuscript as the followings;

Major concerns

1. From table 5, the percentage of combination of anti-platelet and anti-coagulant (warfarin) was obviously higher in PPI group (0% vs. 57%). This maybe the factor contributing higher rebleeding rate in PPI group. Please discuss on this issue.

2. In discussion part, the author stated that “In patients taking antithrombotic drugs, vonoprazan was expected to reduce post-ESD bleeding to a greater extent than PPIs, and the observational study by Kagawa et al. [7] suggested that vonoprazan reduced post-ESD bleeding in patients taking antithrombotic drugs. However, these results were inconsistent with those of the subsequent RCT by Hamada et al. [8], which led us to perform the present study.” In fact, the study by Hamada excluded the patients taking antithrombotic drugs, thus the results were different from previous observational study as they assessed the patients in different group. Please consider re-write.

3. The authors said that “An intragastric pH of 5.4 or higher is required for blood coagulation to achieve haemostasis in peptic ulcer” and showed that Vonoprazan provided the targeted pH. However, this was not only the ability of Vonoprazan but also PPIs. Please revised those sentences, otherwise the reader maybe confused.

4. Please put the reference to this sentence “It remains unanswered whether the results obtained in RCTs are inconsistent with the facts observed in real-world clinical practice, which involves treating patients from diverse backgrounds.” In theory, this sentence may not true as the real-world data contains various type of bias and that is the reason for conducting RCT.

5. The discussion part was too long and some sentences eg. “Keith and Constance reported that clinical studies evaluating the effects of alcohol treatment excluded patients with more severe alcohol disorders and those with low incomes and psychiatric problems”, etc. were not relevant to the present study. Please consider shortening the discussion part.

Reviewer #2: Hidaka et al. conducted a cross-design synthesis evaluating the treatment effects between Vonoprazan and Proton-pump inhibitor in anti-thrombotic users who underwent gastric ESD. They concluded that from the pooled data of selected from one RCT and one observational data from their center as a historical control, Vonoprazan is more effective than PPI in preventing post-ESD bleeding among patients using anti-thrombotics. The study is clinically relevant with an interesting concept. However, a few shortcomings exist.

1. The definition of “taking anti-thrombotic medication” lacks the timing and duration in relation to the procedure i.e. were the medications taken throughout periprocedural period or were they stopped at any point prior to the procedure.

2. Among 3 RCTs included, the other two studies had sample size calculated too. Please specify more elaborative reasons why the study by Hamada et al. was chosen over the other two.

3. Since Vonoprazan requires shorter onset of action and shorter duration to reach its steady level, administering oral PPI one day prior to the ESD procedures may not be adequate time for bleeding prophylaxis. This limitation should be discussed.

4. As ESD with PPI prophylaxis performed before Vonoprazan became available were used as historical control, the bleeding outcome is inevitably confounded by endoscopists' experience which would be less than ESDs performed later (Vonoprazan group). Please discuss this limitation.

5. Statistically significant difference between each type of antithrombotic medications in each group (PPI vs Vonoprazan) should be presented in Table 5.

6. The authors rightfully stated that the risk of post-ESD bleeding differed between individual DOACs and other antithrombotic medication therefore, the differences between each antithrombotic regimen in both Vonoprazan and PPI group in this study needs to be further elaborated.

7. There is a few numerical errors e.g. Specific anti-thrombotic stratification is presented in Table 5, not 4

8. The difference in pharmacokinetic of vonoprazan and PPI should be moved to introduction

9. There are a few grammatical and spelling errors. Please proofread the manuscript again.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

3 Nov 2021

We thank the reviewers for their careful reading of our manuscript and their constructive criticism. The response to the reviewer has been written in the reviewer to response file.

Submitted filename: Response_to_Reviewers.docx

Click here for additional data file.

9 Dec 2021

Efficacy of vonoprazan against bleeding from endoscopic submucosal dissection-induced gastric ulcers under antithrombotic medication: A cross-design synthesis of randomized and observational studies

PONE-D-21-19373R1

Dear Dr. Hidaka,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Sanjiv Mahadeva, MRCP, MD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The revised manuscript is satisfactory

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors did response to the comments properly. The manuscript is well written. I have no further comments for this article.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

14 Dec 2021

PONE-D-21-19373R1

Efficacy of vonoprazan against bleeding from endoscopic submucosal dissection-induced gastric ulcers under antithrombotic medication: A cross-design synthesis of randomized and observational studies

Dear Dr. Hidaka:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Sanjiv Mahadeva

Academic Editor

PLOS ONE