This was a prospective, multicenter, randomized controlled, non-inferior clinical study with the main objective of evaluating the H. pylori eradication rate of rescue therapy containing JWC compared with that of bismuth quadruple therapy.

The study was conducted from January 2019 to December 2022 at Peking University First Hospital, Guang’anmen Hospital of China Academy of Chinese Medical Sciences, Beijing Jishuitan Hospital, Dongzhimen Hospital Beijing University of Chinese Medicine, Dongfang Hospital Beijing University of Chinese Medicine, and Beijing University of Chinese Medicine Third Affiliated Hospital. Patients with H. pylori infection that had failed to be eradicated by quadruple therapy and necessitated rescue treatment were enrolled. Demographic and clinical data of all patients were recorded at the time of enrollment.

The experiment used a random block design. After sequential numbering at each center, patients were randomly assigned in a 1:1:1:1 ratio to a positive control group receiving furazolidone with a bismuth quadruple 14-day regimen and to three treatment groups receiving JWC. During follow-up visits, adverse reactions and medications were recorded. All participants were examined via a urea breath test 4 weeks after the end of eradication therapy to assess H. pylori eradication rates. The symptoms of patients were assessed at baseline, at the end of the treatment period, and 1 month, 3 months, and 6 months after the end of treatment.

The inclusion criteria were as follows: (1) patients who ranged in age from 18 to 70 years old; (2) patients who had previously experienced treatment failure 1–3 times under quadruple therapy; (3) patients who had discontinued treatment for at least 3 months; (4) furazolidone has not been administered in prior therapeutic protocols; (5) patients who had undergone gastroscopy or imaging examinations within the past 2 years to exclude malignant gastric lesions were included.; (6) patients who had a current H. pylori infection; (7) patients who agreed to sign the informed consent forms.

The exclusion criteria were as follows: (1) constant use of antibiotics (against infection under any circumstances), bismuth, PPIs or H2-receptor antagonists within 4 weeks before the H. pylori assessment; (2) history of peripheral neuropathy; (3) allergies to the drug used in this study; (4) pregnancy and lactation; (5) participation in other clinical trials during the previous 3 months; (6) severe or unstable cardiopulmonary or endocrine diseases or severe substantial organ impairments and complications; (7) inability to express oneself properly to cooperate.

Stop/exit/exclusion criteria: (1) deterioration or severe complications; (2) severe adverse reactions occur during treatment and cannot be tolerated; (3) other diseases interfered with observation during treatment; (4) The patient does not cooperate and changes the medication regimen at will; (5) lost to follow-up; (6) pregnancy during treatment; (7) The patients voluntarily asked to withdraw from the study; (8) Any situation or condition that puts the subjects at significant risk.

Eligible patients at each center were randomly divided into four groups, and the intervention regimens were as follows: (i) control group: rabeprazole (Jumpcan Pharma, Taixing, Jiangsu) 20 mg, amoxicillin (Zhuhai United Laboratories Co) 1,000 mg, bismuth potassium citrate (Livzon Group Livzon Pharmaceutical Factory, Zhuhai, Guangdong) 220 mg and furazolidone (Chifeng Mengxin Pharm, Chifeng, Neimenggu) 100 mg twice a day (b.i.d.) for 14 days; (ii) group A: JWC (Tianjin Tasly Pharm, Tianjing) 240 mg b.i.d. was taken instead of bismuth in the control group for 14 days; (iii) group B: receiving the same dose as group A for the first 14 days plus JWC 240 mg b.i.d. for another 14 days; and (iv) group C: amoxicillin-furazolidone quadruple therapy and JWC 240 mg b.i.d. were used together for 10 days. Amoxicillin and furazolidone were taken after meals, while other drugs were taken before meals.

Detection of H. pylori infection was completed by the ¹³C test (¹³C-UBT), rapid urease test (RUT), H. pylori stool antigen test (HpSAT), or histological examination. At least one positive result for the above tests was confirmed as a present H. pylori infection, and a borderline result for ¹³C (DOB:4 ± 2‰) needed to be rechecked.

The ¹³C-UBT was rechecked at least 1 month after the last treatment, and a negative result was defined as H. pylori eradication. The diagnostic threshold for a negative ¹³C-UBT result was established at DOB < 4‰, consistent with current international guidelines. The borderline result for ¹³C-UBT (DOB:4 ± 2‰) needed to be rechecked. Patients who have successfully eradicated H. pylori should undergo a repeat H. pylori test 6 months after the completion of eradication therapy.

Subjects were informed of possible adverse drug reactions before taking the drugs and were asked to record any abnormal sensations on the diary card. Adverse reactions and serious adverse events: Special attention should be paid to furazolidone-related adverse reactions, including dizziness, nausea, rash, fever, polyneuritis, and even psychiatric disorders during medication. All adverse events (AEs) were documented from occurrence to resolution throughout the study period. Serious adverse events (SAEs) were defined as any of the following outcomes attributed to the investigational drug: death, permanent or severe disability, functional impairment, hospitalization, or life-threatening conditions. All SAEs occurring during the trial were reported to the Medical Ethics Committee of Peking University First Hospital and the National Medical Products Administration (NMPA) within 24 h. A standardized SAE report form was completed, and relevant stakeholders (e.g., sponsors, regulatory authorities) were notified immediately. In emergency situations, investigators promptly implemented appropriate management measures. Adverse Reaction Incidence Rate was calculated as: Incidence (%) = (Number of participants with adverse reactions/Total number of participants receiving medication) × 100.

The clinical symptom score was evaluated by the symptom severity index. Symptom information was recorded at baseline, after the end of eradication therapy, and 4 weeks after the end of eradication therapy. The quantitative criteria of symptom grading were evaluated according to the “guiding principles for Clinical Research of Piman syndrome (chronic gastritis and dyspepsia)” in the “Guiding Principles for Clinical Research of New Chinese Medicine” (26). The severity and frequency of clinical symptoms were divided into none, mild, moderate and severe, and the corresponding scores were 0, 1, 2, and 3. Symptom scores were calculated as follows: symptom score = symptom degree score × frequency score. The total symptom score was calculated as the sum of the scores for all symptoms.

Each participant was provided with a paper medication diary card and a caution card to remind them not to miss or take the wrong medication (Supplementary Table 2). During the medication period, the research assistant actively and timely solved the patients’ questions and informed the patients to review to reduce the dropout rate and loss to follow-up rate. Poor compliance was defined as taking <80% of pills.

The primary outcome of the study was the H. pylori eradication rate 4 weeks after the end of treatment. Secondary outcomes included the rate of clinical symptom improvement (at the end of treatment and 4 weeks after the end of treatment), the rate of adverse events and the recurrence rate after eradication.

Due to the influence of COVID-19, some patients could not be reexamined at 4 weeks after the end of treatment, and the results of the breath test within 6 months after eradication treatment were finally determined.

According to reports and previous studies (27, 28), furazolidone bismuth-containing quadruple therapy for 14 days demonstrated a high H. pylori eradication rate exceeding 90%. We assumed that the effective rate of each treatment group was the same as that of furazolidone bismuth-containing quadruple therapy for 14 days. In this study, each group receiving JWC was compared with the control group, with α = 0.05/3 = 0.017 and a power of 0.8. According to the clinically acceptable noninferiority level, the noninferiority margin value was −0.1, and the parallel control between the groups was designed 1:1:1:1. The sample size of the experimental group was 110 cases calculated by PASS 11.0 software. At a dropout rate of 10%, 122 patients per group were needed, resulting in a final requirement of 488 patients.

All statistical analyses were performed using IBM SPSS statistical software version 25.0 (IBM Inc., New York, USA) and SAS software, version 9.4 (SAS Institute Inc., Cary, NC, USA). Measurement data with a normal distribution are expressed herein as the mean ± standard deviation, and analysis of variance was used for comparisons between multiple groups of data. Measurement data with a nonnormal distribution were expressed asdescribed by the median (lower quartile, upper quartile). Count data and rank data were expressed as the number of cases and percentage. The chi-square test or Fisher’s exact test was used for the comparison of count data between groups, and the Kruskal–Wallis test was used for the comparison of multiple groups of rank data. The H. pylori eradication rate and the hypothesis of noninferiority were assessed based on ITT (including the participants who were enrolled in the study), MITT(including the participants who took at least one dose of medicine and underwent the endpoint measure), and PP(including the participants who were fully adherent with the protocol and excluding those with poor compliance) analyses. All tests were two-sided, and p < 0.05 was considered statistically significant.

There were no significant differences in the baseline characteristics, such as demographics, family history, number of eradications, and history of antibiotic treatment, among the four groups (all p > 0.05) (see Table 1).

A total of 433 patients underwent postmedication reexamination, with 414 patients reexamined at 4 weeks, 5 patients at 3 months, and the remaining patients at 6 months. ITT analysis showed that the eradication rate was 85.2% (104/122) in the control group and 73.8% (90/122), 78.7% (96/122) and 75.4% (92/122) in the treatment groups. In the MITT analysis, the eradication rates of the four groups were 92.0% (104/113), 84.9% (90/106), 88.9% (96/108), and 86.8% (92/106). PP analysis showed that the eradication rate was 92.5% (99/107) in the control group, 85.4% (88/103) in group A, 87.9% (94/107) in group B, and 86.7% (91/105) in group C. No statistically significant differences among the groups were found (p = 0.136, p = 0.398 and p = 0.405 for the ITT, MITT, and PP analyses, respectively), which suggests that the eradication rates of the three treatment groups are noninferior to the rate of the control group. The detailed results are shown in Table 2. Additionally, no statistically significant differences in eradication rates were observed between each of the three treatment groups and the control group (all p > 0.05), as detailed in Supplementary Tables 3, 4.

Non-inferiority analysis of H. pylori eradication rates, based on MITT analysis, were performed for each intervention group versus the control group. As summarized in Table 3, group B demonstrated non-inferiority with an eradication rate of 88.9% (p = 0.0415; 90% CI, −0.0965 to 0.0336). In contrast, group A exhibited an eradication rate of 84.9% (p = 0.2527; 90% CI, −0.1422 to 0.0004), and group C achieved an eradication rate of 86.8% (p = 0.1264; 90% CI, −0.1208 to 0.0160), neither of which met the prespecified non-inferiority margin.

Among the 488 patients, 30 patients had adverse events, including 12 cases (9.84%, 12/122) in the control group, 7 cases (5.74%, 7/122) in group A, 8 cases (6.56%, 8/122) in group B, and 3 cases (2.46%, 3/122) in group C. There was no significant difference in the incidence of adverse events among the four groups (p > 0.05). The adverse reactions of the four groups all disappeared after drug withdrawal and other corresponding treatments. Details of the adverse reactions in 30 patients are shown in Table 4.

As shown in Table 5, the four groups of patients with symptoms showed 69.8, 74.5, 80.9, and 86.3% improved effectiveness, respectively, at the end of treatment, and the differences among the four groups were statistically significant (p < 0.05). After 4 weeks of treatment, the effective improvement rates of symptoms in each group were 76.0, 85.7, 87.2, and 87.1%, respectively, which were not significantly different among the four groups (p > 0.05). There was also no significant difference in medication compliance (p > 0.05).

We also compared improvements in major individual digestive symptoms (see Table 6). At the end of medication, the patients’ overall symptoms scores decreased. Among them, the curative effect of groups A, B and C was better than that of the control group in terms of bitter taste and dry mouth and abdominal pain, and the difference was statistically significant (p < 0.05). It can also be seen from Table 7 that four weeks after the end of medication, the symptoms of bitter taste and dry mouth in patients in groups A, B and C improved more significantly than those in the control group, and the symptoms of gastric noise of patients in groups B and C improved significantly, all of which were statistically significant (p < 0.05).

Some patients had no clinical symptoms before treatment, at the time of drug withdrawal, and 4 weeks after drug withdrawal, so they were not included in the calculation of the response rate. To ensure data integrity, we implemented online symptom inquiries for patients who were unable to attend follow-up appointments due to COVID-19.

The follow-up for recurrence rates was conducted only with patients who tested negative on the breath test during reexamination of eradication 4 weeks after completing treatment. The H. pylori reinfection rate was higher in the control group than in the other three groups at 6 months after successful eradication, but the differences were not statistically significant (p > 0.05) (see Supplementary Table 1).