Gastroenterology · Medical Literature
Vonoprazan-based dual therapy with amoxicillin or doxycycline versus bismuth-containing quadruple therapy for Helicobacter pylori eradication: a prospective, multicenter, open-label, randomized controlled trial.
Abstract
Bismuth-containing quadruple therapy, the standard Helicobacter pylori (H. pylori) eradication regimen, is limited by adverse events and rising resistance. Vonoprazan-amoxicillin dual therapy offers simplicity but has mostly been tested using high-dose amoxicillin. Data on doxycycline in this context are scarce, and vonoprazan-doxycycline dual therapy remains unexplored. A total of 579 H. pylori-positive patients were enrolled and randomly assigned in a 1:1:1 ratio to one of three 14-day treatment groups: the VBQ group (vonoprazan 20 mg, amoxicillin 1000 mg, doxycycline 100 mg, and colloidal bismuth pectin 300 mg, all twice daily); the VA group (vonoprazan 20 mg and amoxicillin 1000 mg, twice daily); and the VD group (vonoprazan 20 mg and doxycycline 100 mg, twice daily). A 13C-urea breath test (UBT) was performed at least four weeks after treatment completion. H. pylori eradication rates, the incidence of adverse events during treatment, and medication adherence were compared among the three groups. In the intention-to-treat (ITT), modified intention-to-treat (mITT), and per-protocol (PP) analyses, the VA regimen demonstrated non-inferiority to the VBQ regimen. The VD regimen did not meet the non-inferiority criterion in the ITT analysis (79.3% vs. 83.4%, P = 0.070 for non-inferiority), but it demonstrated non-inferiority to the VBQ regimen in both the mITT (87.4% vs. 90.4%, P = 0.018) and PP (87.7% vs. 90.1%, P = 0.012) analyses. The overall incidence of adverse events was not significantly different between the VBQ and VA groups (P = 0.313), but it was significantly lower in the VD group compared to the VBQ group (P = 0.002). No statistically significant differences in medication adherence were observed between the VBQ group and either the VA (P = 0.381) or VD (P = 0.269) group. Vonoprazan-low-dose amoxicillin dual therapy is a safe and simple alternative to bismuth containing quadruple therapy. Vonoprazan-doxycycline dual therapy demonstrates acceptable efficacy and may be considered for patients allergic to amoxicillin. Both align with antimicrobial stewardship principles of regimen simplification.
Clinical relevance
In H. pylori-positive patients, is vonoprazan-amoxicillin (VA) or vonoprazan-doxycycline (VD) dual therapy non-inferior to bismuth-containing quadruple therapy (VBQ) for eradication at 14 days, while offering a comparable or improved safety profile and adherence? __SNAPSHOT__ {"studyDesign":"Prospective, multicenter, open-label, three-arm, non-inferiority randomized controlled trial","setting":"Multiple centers in China (lead center: Department of Gastroenterology, Nanjing First Hospital, Nanjing Medical University, Jiangsu); specific number of sites not reported in the abstract","sampleSize":"n = 579, randomized 1:1:1 — VBQ group (vonoprazan + amoxicillin + doxycycline + colloidal bismuth pectin, n ≈ 193), VA group (vonoprazan + amoxicillin, n ≈ 193), VD group (vonoprazan + doxycycline, n ≈ 193)","population":"H. pylori-positive patients; included both treatment-naïve and prior eradication failure (rescue therapy) patients; specific inclusion/exclusion criteria beyond H. pylori positivity not reported in the available abstract","intervention":"VA group: vonoprazan 20 mg + amoxicillin 1000 mg, both twice daily for 14 days (low-dose amoxicillin dual therapy); VD group: vonoprazan 20 mg + doxycycline 100 mg, both twice daily for 14 days (novel doxycycline dual therapy)","comparator":"VBQ group (active control): vonoprazan 20 mg + amoxicillin 1000 mg + doxycycline 100 mg + colloidal bismuth pectin 300 mg, all twice daily for 14 days","primaryOutcome":"H. pylori eradication rate assessed by 13C-urea breath test (UBT) at least 4 weeks after treatment completion; non-inferiority margin prespecified at −10% (lower limit of 95% CI for risk difference must be above −10%)","followUp":"Minimum 4 weeks post-treatment (UBT confirmation); no long-term follow-up reported","fundingSource":"Not reported"}
Key findings
The primary analysis was the modified ITT (mITT) population. For the VA vs. VBQ comparison, the mITT eradication rates were 90.3% (VA) vs. 90.4% (VBQ), with a risk difference of −0.11% (95% CI: −6.25% to 6.03%), and a non-inferiority P-value <0.001; no statistically significant difference between regimens (P = 0.972). Non-inferiority of VA to VBQ was also established in the ITT analysis (82.4% vs. 83.4%, P = 0.010 for non-inferiority) and the PP analysis (90.5% vs. 90.1%, P <0.001 for non-inferiority). For the VD vs. VBQ comparison, the mITT analysis showed eradication rates of 87.4% (VD) vs. 90.4% (VBQ), with non-inferiority achieved (P = 0.018); PP analysis similarly demonstrated non-inferiority (87.7% vs. 90.1%, P = 0.012). However, VD failed to achieve non-inferiority in the ITT analysis (79.3% vs. 83.4%, P = 0.070), meaning non-inferiority was not established across all three pre-specified analysis populations. In a subgroup analysis of treatment-naïve patients, the VA vs. VBQ mITT risk difference was 2.16% (95% CI: −4.14% to 8.46%), supporting non-inferiority; in the rescue therapy subgroup, the mITT risk difference for VA vs. VBQ was −37.50% (95% CI: −71.05% to −3.95%), suggesting inferior VA performance as a rescue regimen, though this subgroup was presumably small and the CI is wide. Regarding safety, the overall incidence of adverse events did not differ significantly between the VBQ and VA groups (P = 0.313). The VD group had a significantly lower adverse event incidence compared to VBQ (P = 0.002). No statistically significant differences in medication adherence were observed between VBQ and VA (P = 0.381) or between VBQ and VD (P = 0.269). Absolute adverse event rates and specific adverse event types by arm were not reported in the available full-text excerpts. NNT and NNH cannot be calculated from the available data.
Practice implications
This trial establishes that 14-day vonoprazan–amoxicillin (low-dose, 1 g BID) dual therapy is non-inferior to a vonoprazan-based bismuth-containing quadruple regimen for H. pylori eradication across all three analysis populations (ITT, mITT, and PP), with comparable adverse event rates and adherence. This supports VA dual therapy as a legitimate, simplified alternative to quadruple therapy in patients without penicillin allergy, aligning with antimicrobial stewardship goals of regimen simplification and reduced antibiotic burden. The trial also provides the first RCT evidence for vonoprazan–doxycycline dual therapy, which achieved non-inferiority in the mITT and PP analyses but failed the ITT non-inferiority criterion. Because regulators and guidelines generally require non-inferiority across all analysis populations — particularly ITT — VD cannot be considered formally non-inferior based on this trial alone. Nevertheless, the mITT and PP results are clinically encouraging, and VD may be a reasonable option for patients with documented penicillin allergy when alternatives are limited. An important signal of concern is the rescue therapy subgroup, where VA performed substantially worse than VBQ; this limits the applicability of VA dual therapy in the retreatment setting and requires confirmatory study. Findings are most applicable to the Chinese patient population in which the trial was conducted, where CYP2C19 polymorphism prevalence, antibiotic resistance patterns (particularly low amoxicillin resistance), and local H. pylori strains differ from Western populations. Generalizability to North American, European, or other high-resistance settings is uncertain. The study examined only a treatment and confirmation window; long-term outcomes such as gastric cancer prevention were not assessed. Only the abstract and selected full-text excerpts were available for this review, which limits assessment of methodological details including randomization procedures, allocation concealment, blinding adjudication, full subgroup definitions, and complete adverse event data. __LIMITATIONS__ Open-label design: neither patients nor treating clinicians were blinded to treatment allocation, introducing performance and assessment bias risk, particularly for subjective adverse event reporting.,The VD regimen failed the ITT non-inferiority criterion (P = 0.070), making its non-inferiority claim dependent on mITT and PP analyses alone; ITT is typically the most conservative and regulatory-relevant analysis in non-inferiority trials.,Mixed population of treatment-naïve and rescue therapy patients without pre-stratified power calculations for subgroups; the rescue therapy subgroup for VA appeared to perform substantially worse than VBQ (risk difference −37.50%), but subgroup sizes and statistical power for this comparison are not reported.,No antibiotic susceptibility testing was performed on H. pylori isolates, making it impossible to determine whether eradication failures were attributable to resistance to amoxicillin, doxycycline, or both.,Conducted exclusively in a Chinese population; results may not be generalizable to Western populations with different H. pylori antibiotic resistance profiles, CYP2C19 polymorphism frequencies, and vonoprazan pharmacodynamics.,The VBQ comparator is a vonoprazan-based quadruple regimen rather than the more standard PPI-based bismuth quadruple regimen, limiting direct comparison to standard-of-care in most non-Asian settings.,Funding source is not reported, precluding assessment of potential conflicts of interest.,Specific adverse event rates by type and severity across arms, as well as complete adherence data, were not available in the extracted full-text sections; the full published article should be reviewed for these details.,Follow-up was limited to 4 weeks post-treatment for UBT confirmation; no data on long-term recurrence, reinfection, or impact on downstream gastric cancer risk.,The 14-day amoxicillin dose used (1 g BID = 2 g/day) is lower than the commonly studied high-dose regimen (3 g/day); while this is a stewardship advantage, efficacy in populations with higher amoxicillin MICs is uncertain. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ Fourteen-day vonoprazan–amoxicillin (1 g BID) dual therapy is non-inferior to vonoprazan-based bismuth quadruple therapy for H. pylori eradication in a Chinese multicenter population, with a comparable adverse event burden, supporting its use as a simplified, antibiotic-sparing first-line option in penicillin-tolerant patients. Vonoprazan–doxycycline dual therapy showed numerically lower but potentially non-inferior eradication rates in mITT/PP analyses and a significantly lower adverse event rate, making it a potentially viable option for penicillin-allergic patients, though it failed the ITT non-inferiority test and cannot yet be considered formally established. Clinicians should exercise caution in applying VA dual therapy as rescue therapy given signals of substantially inferior efficacy in previously treated patients.