Gastroesophageal reflux disease (GERD) is a chronic disorder,and although effective short-term treatment strategies are known, the rate of relapse within 1 year is as high as 90% despite successful acute treatment. Consequently, most patients with GERD require an effective long-term management strategy to achieve adequate symptom control and maintain mucosal healing.
The present study was undertaken to compare the control ofGERD symptoms during long-term (24-week) treatment with pantoprazole 20 mg used on-demand or continuously in patients with mild GERD after complete relief of acute GERD symptoms.
Patients with endoscopically confirmed Savary/Miller grade 0(normal mucosa) or I (patchy red lesions without white coating or with central white coating) GERD were enrolled in this multinational, multicenter study comprising 2 phases. In the first phase, which was open label, patients were treated with pantoprazole 20 mg QD for 4 weeks. The presence and intensity of the symptoms of heartburn, acid regurgitation, and pain on swallowing were assessed. In the second phase, which was an open-label, 24-week, randomized design, only patients completely free of GERD symptoms after acute treatment were included. During this phase, on-demand treatment with pantoprazole 20 mg was directly compared with continuous treatment. The rate of failure to control GERD symptoms after 24 weeks of treatment was estimated using the Kaplan-Meier method. Subsequently, the difference between treatments (on-demand minus continuous) and its 95% CI were calculated, and the on-demand treatment was tested for noninferiority using a predefined noninferiority margin of 20%. The mean daily symptom loads were compared between the treatment groups using the 1-sided Wilcoxon rank sum test on a 5% α level. The point estimate of the difference was determined using the Hodges-Lehman estimator and the 1-sided 95% CI according to Moses. The number of patients unwilling to continue due to insufficient control of heartburn, acid regurgitation, and pain on swallowing was analyzed using the Kaplan-Meier (time-to-event) analysis. Analysis was performed in the same manner as for the rate of failure to control GERD symptoms, but the 95% CI was interpreted for statistical superiority.
A total of 558 patients were enrolled in this study. At the end of theacute phase, 82.1% of patients in the per-protocol (PP) population and 79.1% in the intent-to-treat (ITT) population were relieved of all GERD symptoms, and subsequently entered the long-term phase. After 24 weeks of treatment, analysis of the failure rates revealed that on-demand treatment was noninferior to continuous treatment because the 95% CI was completely below 20% (ITT, 12.1% difference [95% CI, −∞ to 18.9%]; PP, 10.1% difference [95% CI, −∞ to 17.7%]). The higher perceived mean (SD) daily symptom load in the on-demand group (ITT, 1.26 [1.491 vs 0.82 [1.341) was balanced by the reduced tablet intake in that group (PP, 0.51 [0.31 ] vs 0.97 [0.11 ] tablets/d; P < 0.001). With respect to the rate of patients unwilling to continue treatment, no statistically significant difference was observed between the on-demand and continuous groups (ITT/PP, 0.95/1.13 vs 0.95/1.26).
In this study of pantoprazole 20 mg tablets in patients withmild GERD, patients receiving on-demand treatment benefited despite their higher symptom load. The similar rates of unwillingness to continue treatment in both groups might suggest that patients were satisfied with the on-demand treatment strategy. On-demand treatment with pantoprazole 20 mg was found to be noninferior compared with continuous therapy with regard to symptom control. Both on-demand and continuous treatments were well tolerated.