BACKGROUND: Erosive esophagitis (EE) is commonly managed with proton pump inhibitors (PPIs), yet many patients experience incomplete healing or recurrence. Potassium-competitive acid blockers (P-CABs) have emerged as potential alternatives, but high-certainty comparative evidence across agents remains limited. We performed a network meta-analysis to evaluate the relative efficacy and safety of P-CABs versus PPIs and to assess the certainty of the evidence.

METHODS: We systematically searched PubMed, the Cochrane Library, and Web of Science from inception through March 1, 2025, for randomized controlled trials (RCTs) comparing P-CAB, PPI, and/or placebo for the treatment of EE. Risk of bias was assessed using the Cochrane Risk of Bias 2.0 tool. Key outcomes were 8-week endoscopic healing and 24-week recurrence. Certainty of evidence was evaluated using GRADE. Risk difference (RD) estimates were calculated using random-effects models. The study protocol was registered with PROSPERO (CRD420251116179).

FINDINGS: Thirty-nine RCTs were included; all evaluated once-daily dosing. At 8 weeks, zastaprazan 20 mg, vonoprazan 20 mg, and esomeprazole 40 mg demonstrated moderate-certainty superiority over rabeprazole 20 mg and omeprazole 20 mg with RDs ranging from 0.05 to 0.11, while only vonoprazan 20 mg demonstrated moderate-certainty benefit versus lansoprazole 30 mg (RD: 0.04). In Los Angeles (LA) grade C/D EE, vonoprazan 20 mg, esomeprazole 40 mg, and rabeprazole-ER 50 mg demonstrated moderate-to-high-certainty benefit over lansoprazole 30 mg and omeprazole 20 mg with RDs ranging from 0.05 to 0.15. Vonoprazan 20 mg and rabeprazole-ER 50 mg demonstrated moderate-certainty benefit compared with pantoprazole 40 mg (RDs: 0.12 and 0.09, respectively).At 24 weeks, vonoprazan 10 mg and 20 mg showed moderate-to-high-certainty benefit versus lansoprazole 15 mg (RDs: -0.11 and -0.13, respectively), while in direct comparisons, esomeprazole 20 mg outperformed lansoprazole 15 mg and pantoprazole 20 mg, with approximately 40-50% relative reductions in recurrence. In LA grade C/D EE, vonoprazan 10 mg and 20 mg demonstrated moderate-to-high-certainty superiority over lansoprazole 15 mg and pantoprazole 20 mg with RDs ranging from -0.12 to -0.20. Esomeprazole 20 mg showed a high-certainty benefit compared with pantoprazole 20 mg (RD: -0.16). At 8 and 24 weeks, safety profiles were generally comparable between P-CABs and PPIs.

INTERPRETATION: Among once-daily regimens, vonoprazan 20 mg, zastaprazan 20 mg, and esomeprazole 40 mg were most effective for healing EE, while vonoprazan 10 mg and 20 mg and esomeprazole 20 mg were most effective in preventing recurrence. Benefits were most pronounced in LA grade C/D EE and are supported by moderate to high-certainty evidence. Comparative trials evaluating newer P-CABs against optimized PPI strategies, including twice-daily dosing, are needed to evaluate efficacy and long-term safety, particularly with respect to hypergastrinemia and infection risk.

FUNDING: None was received for the study.