Long-term acid blocker safety review finds hypergastrinemia but no clear cancer signal

Potassium-competitive acid blockers, or P-CABs, have rapidly become some of the most potent heartburn and ulcer drugs on the market, but what happens when patients take them for years rather than weeks? A new narrative review published in Frontiers in Pharmacology on 28 October 2026 pulls together the available long-term safety evidence and reaches a cautiously reassuring conclusion: the best-studied agent, vonoprazan, has not shown a substantially higher overall incidence of adverse events than proton pump inhibitors, the older drug class it increasingly competes with. Yet the review also flags persistent gaps, because the evidence base is dominated by a single drug, and the risks of rare or slow-developing outcomes remain genuinely unknown.

P-CABs include vonoprazan, tegoprazan, keverprazan, fexuprazan, and zastaprazan. Unlike proton pump inhibitors, which irreversibly disable the stomach’s acid-producing pump, P-CABs bind reversibly to the potassium-binding site of gastric H+/K+-ATPase. The result is a rapid onset of action, potent and sustained acid suppression, and relatively little influence from CYP2C19 metabolizer status, a genetic variation that shapes how patients respond to many proton pump inhibitors. Guidelines have increasingly incorporated these agents into treatment for reflux esophagitis, peptic ulcer bleeding, refractory reflux, and Helicobacter pylori eradication, and as maintenance therapy expands, so does concern about what years of near-total acid suppression do to the stomach.

The review, led by Yuqi Bao and colleagues at Zhejiang Chinese Medical University, searched PubMed, Embase, the Cochrane Library, and ClinicalTrials.gov through 30 June 2026. The authors defined long-term human evidence as studies with at least 24 weeks of treatment or follow-up, a pragmatic cutoff chosen to capture maintenance and extension studies. They deliberately excluded short-term trials from the core evidence base, noting that P-CAB exposure in eradication regimens lasts only 7 to 14 days and cannot be extrapolated to long-term maintenance. Because the included studies differed so much in drug, population, dose, and outcome definitions, no meta-analysis was performed; the evidence was synthesized narratively, and no formal risk-of-bias assessment was carried out.

The strongest evidence comes from randomized trials and extension studies, most of them conducted in East Asia. In a Japanese maintenance study of 305 patients with healed erosive esophagitis, treatment-emergent adverse events occurred in 76.6% and 78.8% of patients receiving vonoprazan 10 or 20 mg over 52 weeks, with nasopharyngitis the most common complaint. A 24-week trial in patients on long-term NSAIDs found adverse event rates of 84.4% to 88.1% across vonoprazan and lansoprazole groups, with overall safety judged comparable. In a United States and Europe trial of 889 patients in the maintenance phase, adverse events ranged from 50.5% to 56.4%, fractures were rare, and no Clostridioides difficile infection occurred. Trials of tegoprazan and fexuprazan reported broadly similar tolerability to lansoprazole, with mostly mild gastrointestinal symptoms such as diarrhea, nausea, and abdominal discomfort.

The most consistently documented long-term effect is hypergastrinemia. Sustained inhibition of the acid pump raises intragastric pH, which removes the feedback that normally restrains gastrin release. Circulating gastrin then stimulates histamine release from enterochromaffin-like cells and acts on parietal cells, but because the pump itself is blocked, these compensatory pathways cannot restore acid secretion. Under prolonged stimulation, the gastric glands remodel. A meta-analysis cited in the review found a greater gastrin increase with P-CABs than with proton pump inhibitors, with a pooled mean difference of 130.92 pg/mL, though the underlying studies were heterogeneous. In one cohort of 48 patients taking vonoprazan for more than a year, gastrin rose sharply early and plateaued after roughly 1.5 years, and severe gastric mucosal atrophy was independently associated with higher levels.

That remodeling is visible through the endoscope. Gastric mucosal redness, characterized histologically by fundic gland dilatation and parietal-cell protrusion, was detected in 8.7% of vonoprazan users in a retrospective study of 1,101 patients, compared with 6.2% for proton pump inhibitor users and under 2% for controls. A distinctive pattern called stardust gastric mucosa, marked by small white elevations from mucus trapped in dilated glands, was found in 4.9% of vonoprazan users versus 0.2% of nonusers after propensity score matching, with cumulative incidence climbing to 26.2% at three years. Web-like mucus appeared in roughly 19% of vonoprazan users in one endoscopy cohort. The review stresses that the clinical significance of these findings remains unclear because bleeding, dysplasia, and tumor progression have not been demonstrated, and that similar changes occur with proton pump inhibitors, suggesting they reflect sustained acid suppression rather than a vonoprazan-specific lesion.

The tumor question is the most charged. In the five-year VISION trial, a phase IV study in which 202 H. pylori-negative patients with healed erosive esophagitis entered maintenance therapy for up to 260 weeks, prespecified gastric biopsies showed no malignant epithelial changes, no gastric neuroendocrine tumors were observed in either group, and ECL-cell hyperplasia was not higher with vonoprazan than with lansoprazole, although parietal-cell and foveolar hyperplasia were more pronounced. One foveolar-type gastric adenoma occurred in each treatment arm. Observational data are more troubling but harder to interpret: a 2024 propensity-matched study of 54,055 Japanese patients after H. pylori eradication found gastric cancer in 1.89% of long-term P-CAB users versus 0.99% of H2-receptor antagonist users, a hazard ratio of 1.92, and a 2026 cohort of patients with previous gastric cancer reported an adjusted hazard ratio of 3.89 for metachronous cancer with P-CAB use. The review cautions that these associations may be affected by residual confounding, differing baseline cancer risk, gastric atrophy, and surveillance frequency, and concludes that current evidence does not demonstrate that P-CABs directly increase gastric cancer or neuroendocrine tumor risk in the general population with acid-related disorders.

Case reports add signal but not proof. One patient who took lansoprazole for 11 years followed by vonoprazan for 3 years developed three grade 1 gastric neuroendocrine tumors with hypergastrinemia and ECL-cell hyperplasia, and another developed a solitary grade 1 tumor after six years of vonoprazan following prior proton pump inhibitor therapy. Because both cases involved sequential acid-suppressive exposure and lacked controls, the review treats them as signal detection only. In rats, high-dose tegoprazan produced gastric ECL-cell tumors at exposures roughly sevenfold above recommended human levels, supporting biological plausibility through the gastrin-ECL axis but not permitting direct extrapolation to clinical doses.

Beyond the stomach, the picture is fragmentary. Observational studies have not shown a consistent excess risk of Clostridioides difficile infection or small-intestinal bacterial overgrowth with P-CABs compared with other acid suppressants, though a TriNetX analysis found a higher 90-day risk of infectious gastroenteritis after vonoprazan initiation than after proton pump inhibitor initiation, a short-term signal that cannot be extrapolated long term. Severe hypomagnesemia has been reported with prolonged vonoprazan therapy, presenting with seizures, hypocalcemia, and heart rhythm disturbances that improved after discontinuation, and the United States prescribing information advises periodic magnesium monitoring during prolonged treatment. Isolated cases of liver injury with fexuprazan, suspected toxic epidermal necrolysis with vonoprazan, and a tacrolimus concentration increase after switching kidney transplant recipients from rabeprazole to vonoprazan round out the extra-gastric signals, most of which rest on single reports or pharmacovigilance databases that cannot estimate incidence.

The review’s bottom line is measured rather than alarming. Long-term safety evidence is derived predominantly from vonoprazan and should not be generalized to the entire class, the authors conclude, with hypergastrinemia and adaptive fundic gland changes the most consistently documented effects. They recommend that long-term prescribing include periodic reassessment of the indication, the lowest effective dose, and individualized monitoring based on patient risk, with particular caution in people who have severe gastric atrophy, intestinal metaplasia, autoimmune gastritis, previous gastric cancer, or hereditary polyposis syndromes. Future research, they argue, should extend beyond vonoprazan to the newer agents, harmonize exposure and outcome definitions, and include populations from diverse backgrounds rather than uncritically extrapolating findings from East Asian settings with high background gastric cancer risk to the rest of the world.

Subject of Research: Long-term safety of potassium-competitive acid blockers

Article Title: Safety of long-term use of potassium-competitive acid blockers: a narrative review

Article References: Bao, Y., Chen, J., Han, J., Le, J., Bao, H., & Li, D. (2026). Safety of long-term use of potassium-competitive acid blockers: a narrative review. Frontiers in Pharmacology, 17, Article 1964932. https://doi.org/10.3389/fphar.2026.1964932

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1964932

Keywords: P-CABs, vonoprazan, hypergastrinemia, gastric mucosal changes, proton pump inhibitors, gastric cancer risk, gastric neuroendocrine tumors, drug safety, acid suppression, narrative review, tegoprazan, fexuprazan