Vonoprazan-amoxicillin dual, rifabutin-based triple, and bismuth quadruple therapies for Helicobacter pylori rescue treatment: a multicentre, open-label, non-inferiority randomised trial

Background Whether vonoprazan-amoxicillin dual therapy (VA) or rifabutin-based triple study (VRA) can replace bismuth quadruple therapy (BQT) for Helicobacter pylori rescue treatment remains unclear. We aimed to evaluate the efficacy and safety of 10-day VA and VRA regimens, compared with BQT, for rescue therapy. Methods We conducted an open-label, non-inferiority randomised controlled trial at three hospitals in China. Refractory H. pylori -infected patients with at least two previous treatment failures were randomly assigned (1:1:1; block sizes six) to 10 days vonoprazan-containing (20 mg twice daily) regimens of VA (amoxicillin 1000 mg three times daily); VRA (amoxicillin 1000 mg three times daily, and rifabutin 150 mg twice daily); or BQT (bismuth 110 mg, tetracycline 500 mg, and metronidazole 400 mg, all four times daily). The primary endpoint was H. pylori eradication rate evaluated by 13 C-urea breath test. Adverse events were assessed in all patients. Non-inferiority margin was −10%. This trial is registered at ClinicalTrials.gov (NCT05874544). Findings Between May 16, 2023, and Jan 31, 2025, 360 subjects (110 [31%] male and 250 [69%] female; mean age 48.4 years [SD 12.1]) were randomly assigned to VA, VRA or BQT (n = 120 per group). Intention-to-treat eradication rates were 75.8% (91/120; 95% CI 67.0–83.0) with VA, 90.0% (108/120; 95% CI 82.8–94.5) with VRA, and 90.8% (109/120; 95% CI 83.8–95.1) with BQT. Non-inferiority of VA to BQT was not established (difference −15.0%; 95% CI −24.2 to −5.8; p=0.86), whereas VRA was non-inferior to BQT (difference −0.8%; 95% CI −8.3 to 6.6; p=0.0078). Resistance rates for amoxicillin, rifabutin, tetracycline, and metronidazole were 10%,