Dual rescue therapy matches cure, fewer side effects
Vonoprazan plus tetracycline cleared Helicobacter pylori as well as standard bismuth four-drug rescue therapy, with far fewer side effects and better completion.
*Open-label randomized noninferiority trial; Level 1b (OCEBM).

Citation

Gao W, Li J, Yang G, et al. Vonoprazan-Tetracycline Dual Regimen as Rescue Therapy for Helicobacter pylori Infection: Randomized Controlled Trial. Gastroenterology. 2026;170:1473–1483. doi:10.1053/j.gastro.2026.01.008.

Background

After a failed attempt to eliminate Helicobacter pylori, rescue treatments often require multiple drugs and can be hard to tolerate. This trial tested whether a simpler two-drug option works as well as standard rescue therapy.

Patients

Adults aged 18–80 years with Helicobacter pylori confirmed by carbon-13 urea breath test and at least 1 prior eradication failure. Exclusions included recent acid-suppressing drugs, antibiotics, or bismuth (within 4 weeks), pregnancy or breastfeeding, severe major illness, or allergy to study drugs.

Intervention

Vonoprazan 20 mg twice daily plus tetracycline 500 mg three times daily for 14 days.

Control

Lansoprazole 30 mg twice daily plus bismuth 150 mg three times daily plus tetracycline 500 mg three times daily plus metronidazole 400 mg three times daily for 14 days.

Outcome

Primary: eradication on carbon-13 urea breath test. Secondary: side effects, stopping treatment because of side effects, and taking at least 80% of doses.

Follow-up Period

Breath test at 8-week follow-up (about 6 weeks after finishing therapy).

Results

Outcome Key finding
Eradication (primary) 90.6% vs 89.3% (difference 1.2 percentage points; 95% range −5.7 to 8.2); met “not worse by >10 points” criterion.
Any treatment-related side effect 10.9% vs 45.7%; 34.8 fewer per 100 treated (number needed to treat: 3 to prevent 1 side effect).
Stopped treatment because of side effects 0.0% vs 8.6% (number needed to treat: 12 to prevent 1 stop).
Took at least 80% of doses 96.0% vs 87.4% (number needed to treat: 12 for 1 additional patient to complete adequately).
Eradication results were reported for patients who took at least one dose and returned for breath testing. This was a noninferiority trial with a prespecified 10 percentage-point margin.

Limitations

Single-center and open-label design may limit generalizability and may bias side-effect reporting. No antibiotic resistance testing was done. The 10-point noninferiority margin may be considered wide.

Funding

Beijing municipal science foundation; Peking University First Hospital youth clinical research grant.

Clinical Application

For rescue treatment after prior failure, consider 14-day vonoprazan plus tetracycline to reduce side effects and improve completion without lowering eradication success.