Introduction

Helicobacter pylori (H. pylori) infection and its related diseases (gastric cancer, peptic ulcer disease, etc.) are still important global health issues.[1,2] With the rapid and significant increase in antibiotic resistance, the efficacies of commonly used eradication regimens for H. pylori infection have decreased significantly.[3,4] Classic bismuth quadruple therapy (BQT), namely bismuth, proton pump inhibitor (PPI), metronidazole, and tetracycline, is currently the most important first-line eradication regimen and recommended by expert consensus and guidelines worldwide.[1,5–8] However, in many areas, tetracycline is difficult to obtain clinically, the incidence of adverse reactions is high, and its use is more complicated, which greatly limits the clinical application of BQT.[9,10] Whether other drugs can be used to replace tetracycline to effectively and conveniently eradicate H. pylori infection has become an important focus.

Minocycline, a type of semisynthetic tetracycline,[11,12] has a better bactericidal activity than tetracycline in many pathogenic bacteria.[13] Owing to the higher lipid solubility, minocycline has a higher absorption rate.[14,15]Minocycline has a long half-life and needs to be taken orally only once or twice daily, facilitating improvement of patient compliance.[12,15] Significantly, H. pylori was sensitive to minocycline in the antibiotic sensitivity testing in vitro, and its resistance rate was similar with that of tetracycline.[11,16] Besides, secondary resistance did not easily occur in patients who failed eradication therapy.[17] Taken together, these features and results suggested that minocycline might be a potent alternative to tetracycline for eradication therapy of H. pylori infection.

However, the clinical research for eradication of H. pylori infection with minocycline-containing regimens is very limited. In the early years, a few case reports of patients with refractory H. pylori infection have shown that minocycline in combination with other drugs may have good eradication efficacy.[18,19] Recently, through preliminary prospective cohort studies, we have shown that the quadruple therapy with bismuth potassium citrate, esomeprazole, metronidazole, and minocycline for 14 days achieved satisfactory efficacy for first-line H. pylori eradication. The eradication rates in intention-to-treat (ITT) and per-protocol (PP) analyses were 85.5% and 92.6%, respectively.[12] Furthermore, another quadruple therapy with minocycline obtained similar good results (ITT analysis: 87.5% and PP analysis: 92.6%).[11] Both studies also showed that the minocycline-containing BQT had relatively good safety and compliance.[11,12] Therefore, conducting more related research, especially randomized controlled trial (RCT) with classic BQT, to determine the role of minocycline-containing BQT in H. pylori eradication is necessary.

The present study aimed to compare the eradication rate, safety, and compliance between 14-day minocycline- and tetracycline-containing BQT as first-line regimens for eradicating H. pylori infection.

Methods Study population

This study was conducted in a tertiary hospital located in Beijing, China between April 2019 and November 2021. The patients with dyspepsia were recruited and considered eligible for enrollment if they were 18–70 years of age and had H. pylori infection.

We excluded patients (1) who previously underwent eradication treatment for H. pylori infection; (2) who have taken drugs that could influence the study results, including PPIs, potassium-channel acid blockers (P-CABs), H2 receptor blockers, bismuth, and antibiotics in the last 4 weeks; (3) with gastrointestinal malignancy; (4) with Zollinger-Ellison syndrome; (5) who underwent previous gastric or esophageal surgery; (6) with severe concomitant diseases; (7) with known allergies to any study drug; or (8) who are currently pregnant or lactating or with alcohol abuse.

Ethical Approval

The study protocol was approved by the Medical Ethics Committee of Peking University Third Hospital (No. M2019059) and conducted according to the principles of the Declaration of Helsinki and the standards of Good Clinical Practice. All patients voluntarily provided informed consent to participate in the study. The study was registered in the Chinese Clinical Trials Registration (No. ChiCTR1900023646).

Study design

This study was a prospective, open-label, single-center, RCT with the primary objective of verifying whether minocycline-containing BQT is noninferior to tetracycline-containing BQT as the first-line regimen for eradicating H. pylori infection. The participants were enrolled based on their eligibility to the inclusion and exclusion criteria. After enrollment, the demographic data and clinical data were collected. The participants were then randomly assigned to one of the two eradication groups in a 1:1 ratio with a parallel arm design. A computer-generated randomization scheme (SAS version 9.1.3; SAS Institute, Cary, NC, USA) was constructed using a block design (block size of four) provided by an independent statistician and was used to determine treatment allocation. The allocation was concealed by using opaque envelopes, which were in turn opened sequentially by the investigator after each patient was deemed eligible and had provided written informed consent. The participants were asked to come back at 1–3 days after eradication to evaluate the safety profile (incidence of adverse events) and compliance (intake of drugs). Participants underwent a 13C-urea breath test (UBT; UCBT Kit; Atom High Tech, Beijing, China) at 4–8 weeks after eradication to evaluate the therapeutic outcomes. Other drugs that could affect the results were prohibited during the study.

Smoking was defined as consumption of >1 pack of cigarettes/week in the previous 6 months. Alcohol drinking was defined as consumption of >50 g of alcohol/day in the previous 6 months. Patients with duodenal and/or gastric ulcer found in upper endoscopy were diagnosed with peptic ulcer disease, while those without ulcer were considered as having nonulcer dyspepsia. Uninvestigated dyspeptic patients included those with dyspeptic symptoms but did not receive upper endoscopy.

Treatment regimens

Minocycline-containing BQT comprised bismuth potassium citrate 110 mg four times daily (before breakfast/lunch/dinner/bedtime), esomeprazole 20 mg twice daily (before breakfast/dinner), metronidazole 400 mg four times daily (after breakfast/lunch/dinner and before bedtime), and minocycline 100 mg twice daily (after breakfast/dinner) for 14 days.

Tetracycline-containing BQT comprised bismuth potassium citrate/esomeprazole/metronidazole with doses same as above and tetracycline 500 mg four times daily (after breakfast/lunch/dinner and before bedtime) for 14 days. Treatment allocation was not blinded.

The study drug information was as follows: bismuth potassium citrate (bismuth potassium citrate capsules 110 mg/capsule; Group Li Zhu Pharmaceutical Factory, Zhuhai, China), esomeprazole (esomeprazole magnesium enteric-coated tablets 20 mg/tablet; AstraZeneca Pharmaceutical Co., Ltd, Wuxi, China), metronidazole (metronidazole tablets 400 mg/tablet; Yabao Pharmaceutical Group Co., Ltd, Yunzhou, China), tetracycline (tetracycline tablets 250 mg/tablet; Hubei Yuancheng Saichuang Technology Co., Ltd, Wuhan, China), and minocycline (minocycline 50 mg/capsule; Wyeth Pharmaceutical Co., Ltd, Suzhou, China).

Safety and compliance

The medical staff in the gastroenterology unit should thoroughly explain the treatment regimens and associated potential adverse events to all enrolled participants. The participants were given both verbal and written instructions about the importance of taking medications regularly and recommended not to stop medication in the event of mild-to-moderate adverse effects. The participants were advised to call the doctor if they had severe side effects. The participants were asked to return at 1–3 days after eradication to determine the incidence of adverse effects and compliance assessment.

The adverse events were determined by asking open-ended questions using patient self-reports and physical examinations and were grouped into mild (no effect on daily routine), moderate (limited effects on daily routine), severe (marked effects on daily routine and medication discontinuation), and serious (death, hospitalization, disability, or required intervention for permanent damage prevention) types.

Compliance was assessed by pill count, which was either considered good (≥80% of pills taken) or poor (<80%). Individuals with poor compliance were not included in the PP analysis.

H. pylori detection

Before enrollment, the status of H. pylori infection was determined by one of the two methods: (1) positive in 13C-UBT and (2) positive in both rapid urea test (RUT, HPUT-H102, San Qiang Bio & Che, Fujian, China) and histological Warthin–Starry staining by upper endoscopy. Post-treatment H. pylori status was assessed by 13C-UBT at 4–8 weeks after treatment. H. pylori infection was considered eradicated if the result of the 13C-UBT was negative.

A gastric biopsy taken from the antrum was subjected to RUT. If the patient was tested positive by RUT, two mucosal biopsy specimens (one each from the antrum and corpus) were obtained for histological Warthin–Starry staining. Two additional specimens (one each from the antrum and corpus) were obtained for culturing and antimicrobial susceptibility testing of H. pylori.

PPIs, P-CABs, H2-receptor blockers, bismuth salts, and antibiotics were discontinued for at least 4 weeks before 13C-UBT was performed. 13C-UBT was performed after overnight fasting. A baseline breath sample was obtained by blowing through a disposable plastic straw into a 20-mL container, and a capsule containing 75-mg 13C-urea was given to patients with 100-mL water. Another breath sample was collected after 30 min. The test was considered positive if the difference between the baseline and 30 min samples exceeded 4.0 parts/1000 of 13CO2, as analyzed by using a gas chromatography isotope ratio mass spectrometer (GC-IRMS; GIRMS ZC-202, Wan Yi Sci & Tech, Anhui, China).

H. pylori strain culture and antimicrobial susceptibility testing

H. pylori culture was performed in all the patients receiving upper endoscopy. H. pylori strains were isolated and cultured from gastric mucosal samples. In vitro, antibiotic resistance was evaluated with the Epsilometer test (AB Biodisk, Stockholm, Sweden).[20]H. pylori strains with minimal inhibitory concentrations of >0.125 μg/mL,>0.5 μg/mL, >8 μg/mL, >1 μg/mL, >1 μg/mL, and >8 μg/mL showed resistance to amoxicillin, clarithromycin, metronidazole, levofloxacin, tetracycline, and minocycline, respectively.[16,21]

The endoscopists, pathologists, and technicians who performed RUT, 13C-UBT, Warthin–Starry staining, H. pylori strain culture, and antimicrobial susceptibility testing were all blinded to the treatment group allocation.

Statistical analysis

In our previous small-scale pilot trial using tetracycline-containing BQT, the eradication rate in ITT analysis was 86.7%, whereas that of minocycline-containing BQT was 85.5%.[12] Noninferiority test analysis was used by setting the primary endpoint (H. pylori eradication rate) noninferiority margin at 10.0% and considering a drop-out rate of 10%, one-sided α = 0.025, and β = 0.20. Altogether, 434 participants with 217 cases per group should be randomized. Differences and 95% confidence intervals (CIs) in the eradication rate between the two groups were calculated. According to the recognized clinical practice of eradication treatment research of H. pylori infection worldwide, the difference of eradication efficacy was set to 10% to evaluate the two regimens, so the noninferiority (minocycline- vs. tetracycline-containing BQT) would be concluded if the lower limit of the 95% CI was>-10.0% (derived using 10.0% as the noninferiority margin).

The primary outcome variable was the eradication rate with minocycline- and tetracycline-containing BQT using ITT (including the participants who were enrolled in the study) and PP (including the participants who were fully adherent to the protocol and excluding those with poor compliance) analyses. Secondary outcome variables included the incidences of adverse effects and compliance.

Statistical analysis was performed using SPSS for Windows (version 20; IBM Inc., NY, USA). Two-sided P< 0.05 was considered to be statistically significant. Categorical variables were described as percentages or frequencies, while continuous variables were described as means ± standard deviation. The eradication rates and 95% CIs were calculated. The intergroup differences were evaluated using Pearson chi-squared or Fisher's exact test for categorical variables and Student's t-test for continuous variables. Univariate analysis was performed to evaluate significant predictive variables for eradication of H. pylori.

Results Participants' characteristics

Altogether, 434 participants were randomized into minocycline-containing (n =217) and tetracycline-containing (n =217) BQT groups. All participants were included in ITT analysis. After excluding participants who were lost to follow-up and with trial protocol violation, UBT rejection, adverse drug event intolerance, and poor compliance, the remaining participants were included in PP analysis [Figure 1].

Figure 1:

Study flowchart of participants with H. pylori infection. ITT: Intention-to-treat; PP: Per-protocol; H. pylori: Helicobacter pylori; BQT: Bismuth quadruple therapy.

The differences in sex, age, body mass index, diagnosis, smoking, alcohol intake, success rate for H. pylori culture, and resistance rate to antibiotics between the two groups were not significant (all P >0.05, Table 1).

Table 1 - Comparison of baseline characteristics of patients with H. pylori infection in minocycline- or tetracycline-containing BQT groups. Baseline characteristics Minocycline-containing BQT (n = 217) Tetracycline-containing BQT (n = 217) Statistics P values Gender (male:female) 104:113 109:108 0.230* 0.631 Age (years) 42.2 ± 13.0 41.0 ± 12.9 0.968† 0.334 Body mass index (kg/m2) 23.0 ± 2.6 23.1 ± 2.6 0.349† 0.727 Smoking (yes:no) 33:184 37:180 0.273* 0.602 Alcohol drinking (yes:no) 34:183 32:185 0.071* 0.789 Diagnosis (PUD:NUD:UID) 16:58:143 14:67:136 0.957* 0.620 H. pylori detection (13C-UBT:upper endoscopy) 143:74 136:81 0.492* 0.483 Successful H. pylori culture 63/74 (85.1) 71/81 (87.7) 0.210* 0.647 Amoxicillin resistance 3/63 (4.8) 3/71 (4.2) <0.001‡ >0.999 Clarithromycin resistance 22/63 (34.9) 24/71 (33.8) 0.019* 0.892 Levofloxacin resistance 23/63 (36.5) 24/71 (33.8) 0.107* 0.743 Metronidazole resistance 38/63 (60.3) 44/71 (62.0) 0.038* 0.844 Tetracycline resistance 3/63 (4.8) 4/71 (5.6) <0.001‡ >0.999 Minocycline resistance 4/63 (6.3) 5/71 (7.0) <0.001‡ >0.999

Values were shown as mean ± standard deviation, n: n or n/N (%). *Pearson chi-squared test. †Student's t-test. ‡Refers to Fisher's exact test. BQT: Bismuth quadruple therapy; NUD: Nonulcer dyspepsia; PUD: Peptic ulcer disease; UBT: Urea breath test; UID: Uninvestigated dyspepsia. H. pylori: Helicobacter pylori.

Eradication rates

The study's primary objective was to determine the eradication rate of H. pylori infection by minocycline-containing BQT and noninferiority to tetracycline-containing BQT. The results of both ITT and PP analyses showed that the rate difference of the lower limit of the 95% CI was>-10.0% (ITT analysis: 83.4% vs. 82.9%, with a rate difference of 0.5% [95% CI: -6.9% to 7.9%]; PP analysis: 91.7% vs. 92.1%, with a rate difference of -0.4% [95% CI: -5.6% to 6.4%]) [Table 2]. Thus, minocycline-containing BQT was not inferior to tetracycline-containing BQT.

Table 2 - Comparison of eradication rates of H. pylori infection in both BQT groups. Items Minocycline-containing BQT Tetracycline-containing BQT Rate difference (95% CI) χ 2 P values for difference * ITT analysis

181/217

(83.4%, 77.6–88.0%)

180/217

(82.9%, 77.1–87.6%)

0.5%

(-6.9% to 7.9%)

0.016 0.898 PP analysis

177/193

(91.7%, 86.7–95.0%)

176/191

(92.1%, 87.1–95.4%)

-0.4%

(-5.6% to 6.4%)

0.025 0.875

*The P values were two-sided and were for comparing the difference of minocycline- and tetracycline-containing group. BQT: Bismuth quadruple therapy; CI: Confidence interval; ITT: Intention-to-treat; PP: Per-protocol.

Safety and compliance

The list and proportion of adverse effects are shown in Table 3. About 75 (34.9%) and 88 (41.1%) patients in minocycline- and tetracycline-containing BQT groups, respectively, had adverse effects. The majority of adverse effects was mild and moderate. Except for dizziness being more common (16.3% vs. 6.1%, P =0.001) in minocycline-containing BQT group, the incidence and severity of adverse events between the two groups were similar. No serious adverse effects were reported.

Table 3 - Comparison of safety and compliance among patients with H. pylori infection in the two groups. Safety and compliance Minocycline-containing BQT (n = 215) * Tetracycline-containing BQT (n = 214) * χ 2 P values Patients with adverse reactions 75 (34.9) 88 (41.1) 1.771 0.183 Mild 42 (19.5) 38 (17.8) 2.722 0.256 Moderate 22 (10.2) 32 (15.0) Severe 11 (5.1) 18 (8.4) Taste distortion 36 (16.7) 46 (21.5) 1.566 0.211 Nausea 43 (20.0) 52 (24.3) 1.150 0.284 Diarrhea 15 (7.0) 20 (9.3) 0.803 0.370 Anorexia 43 (20.0) 50 (23.4) 0.715 0.398 Abdominal pain and discomfort 17 (7.9) 21 (9.8) 0.483 0.487 Fatigue 25 (11.6) 27 (12.6) 0.098 0.754 Headache 12 (5.6) 14 (6.5) 0.174 0.677 Skin rash 6 (2.8) 6 (2.8) <0.001 0.993 Dizziness 35 (16.3) 13 (6.1) 11.239 0.001 Constipation 6 (2.8) 5 (2.3) 0.089 0.766 Compliance 195 (90.7) 192 (89.7) 0.116 0.733

Values are shown as n (%).*Three patients in each group were lost of follow-up. Since one patient in minocycline-containing was lost of follow-up after completing safety and compliance, thus the evaluations can be conducted. The other five patients were lost of follow-up within two weeks, so safety and compliance evaluations cannot be conducted. Therefore, in the minocycline-containing and tetracycline-containing BQT groups, two and three patients were unable to be evaluated for safety and compliance due to lost of follow-up, respectively. BQT: Bismuth quadruple therapy.

Good compliance was achieved in 195 (90.7%) and 192 (89.7%) in the minocycline- and tetracycline-containing BQT groups, respectively, and the difference was insignificant.

Risk factors for eradication failure

Univariate analysis indicated that the eradication rate was significantly higher in compliant patients than in noncompliant patients in both groups. There was no significant effect of sex, age, body mass index, smoking, alcohol drinking, diagnosis, or antibiotic resistance on the eradication rates in both groups [Table 4].

Table 4 - Univariate analysis showing factors influencing eradication efficacy. Minocycline-containing BQT (n = 203) * Tetracycline-containing BQT (n = 202) * Possible factors Eradication rate χ 2 P values Eradication rate χ 2 P values Gender Male 86/99 (86.9) 1.052 0.305 90/103 (87.4) 0.648 0.421 Female 95/104 (91.3) 90/99 (90.9) Age <35 years 54/63 (85.7) 1.169 0.557 60/68 (88.2) 0.766 0.715 35–50 years 74/82 (90.2) 73/80 (91.3) >50 years 53/58 (91.4) 47/54 (87.0) Body mass index <22 kg/m2 59/71 (83.1) 5.786 0.055 64/75 (85.3) 4.564 0.118 22–25 kg/m2 78/82 (95.1) 63/72 (87.5) >25 kg/m2 44/50 (88.0) 53/55 (96.4) Smoking Yes 25/29 (86.2) 0.306 0.528 29/33 (87.9) 0.062 0.764 No 156/174 (89.7) 151/169 (89.3) Alcohol drinking Yes 30/33 (90.9) 0.124 >0.999 28/31 (90.3) 0.056 >0.999 No 151/170 (88.8) 152/171 (88.9) Diagnosis PUD 13/14 (92.9) 0.423 0.866 12/13 (92.3) 0.124 >0.999 NUD 47/54 (87.0) 53/60 (88.3) UID 121/135 (89.6) 115/12