Erosive esophagitis, caused by gastroesophageal reflux, is a common medical problem. At present, therapy for erosive esophagitis primarily focuses on the pharmacological reduction of gastric acid secretion. Decreasing the acidity of gastric juice ameliorates reflux symptoms and facilitates esophagitis healing.1,2 To date, several studies have indicated that the prospect of healing of esophagitis correlates directly with the potency of a medication’s antisecretory effect.3–5 A comprehensive meta-analysis of 136 randomized controlled trials comprising 35,978 patients with esophagitis reported that the healing rate among patients treated with proton-pump inhibitors (PPIs; 83%) was higher than that with histamine-2 receptor antagonists (52%); of note, both rates were higher than that with placebo (8%).3

A large proportion of erosive esophagitis relapses, and only 10–25% of patients remain in remission after 6 months of stopping therapy.6,7 The relapse rate is driven, in part, by the underlying disease severity. Good evidence reveals that patients with severer esophagitis (Los Angeles [LA] grade C and D) are markedly more likely to relapse than those with mild esophagitis (ie, LA grade A and B).7 Considering the higher risk of relapse, most patients with documented severe esophagitis warrant long-term maintenance therapy. However, potential risks of long-term use of PPIs comprise secondary hypergastrinemia, malabsorption, and hypochlorhydria.8 In addition, population-based, epidemiological studies have reported that the long-term use of PPIs correlates with an elevated risk of hip fracture by a factor of 1.4 in patients aged >50 years,9 an increase in the risk of infectious gastroenteritis by a factor of 1.5,10 and a doubling of the risk of Clostridium difficile colitis.11

A cost-effective analysis of the management strategies of erosive esophagitis established that initial therapy with PPIs followed by the on-demand therapy is the most cost-effective approach.12,13 Nevertheless, the optimal duration of initial therapy remains unclear. Tytgat reported that the initial treatment of erosive esophagitis is the standard dose of PPI once a day for 3 months.14 Hsu et al reported that extending PPI therapy from 4 to 8 weeks reduced symptom relapse of GERD.15 Reportedly, both dexlansoprazole and esomeprazole are potent PPIs for gastric acid suppression with excellent symptom relief for patients with GERD.16–19 The benefit of dexlansoprazole MR (Takeda Pharmaceuticals, Osaka, Japan) is that it uses a novel approach through which its dual-delayed release (DDR) formulation extends the plasma concentration and, eventually, extends the acid suppression duration.16

To date, no study has directly investigated the 24-week clinical effects and timing to symptom relief of GERD between 60-mg dexlansoprazole and 40-mg esomeprazole. Hence, this open-label, randomized controlled trial aims to compare the 24-week clinical effects of single doses of dexlansoprazole (60 mg) and esomeprazole (40 mg) with initial 8-week treatment and shift to the on-demand use for patients with LA grades A and B erosive esophagitis.

Between January 2015 and March 2017, we enrolled 86 patients with endoscopy-confirmed LA grade A or B erosive esophagitis who started an 8-week course of PPIs for initial therapy for randomization after excluding 77 patients who declined to participate or fulfilled the exclusion criteria. Of these, 43 patients received the esomeprazole treatment, and 43 patients received dexlansoprazole treatment. As 5 patients were lost to follow-up, 81 patients were examined using the PP analysis (esomeprazole group, 41; dexlansoprazole group, 40; Figure 1). The baseline characteristics both groups were similar in age, sex, diet habits, BMI, and GERDQ score. The number of patients with LA grade B erosive esophagitis was higher in the dexlansoprazole group compared with the esomeprazole group, which was statistically significant (10 vs 3; P=0.030; Table 1). The primary outcome of SHE for the dexlansoprazole and esomeprazole groups were 79.1% (34/43, 95% CI=64.0–90.0%) vs 76.7% (33/43, 95% CI=61.3–88.2%, P=0.795) in the intention-to-treat analysis; and 85.0% (34/40, 95% CI=70.2–94.3%) vs 80.5% (33/41, 95% CI=65.2–91.2%, P=0.591) in the per-protocol analysis, which did not exhibit differences between the two groups. The secondary outcomes, such as CSR, symptom relapse rate, days to symptom resolved, treatment failure rate, and the number of tablets taken in 24 weeks, exhibited no statistical differences between the two groups. However, the esomeprazole group had more days with reflux symptoms in the study period than the dexlansoprazole group (53.9±54.2 vs 37.3±37.8; P=0.008; Table 2).Table 1

The baseline characteristics of patients

In the serial follow-up for the GERDQ score posttreatment, the scores in the baseline, 4, 8, 12, 16, 20, and 24 weeks after treatment for the dexlansoprazole versus esomeprazole groups were 23.2±3.7 vs 23.7±4.7 (P=0.878), 17.1±3.7 vs 18.0±4.1 (P=0.209), 16.4±3.6 vs 16.9±3.7 (P=0.686), 16.3±4.0 vs 17.4±4.7 (P=0.124), 14.7±4.4 vs 16.2±4.7 (P=0.365), 13.7±3.2 vs 15.0±4.8 (P=0.124), and 13.1±3.8 vs 16.5±10.9 (P=0.252), respectively (Table 3). The comparison of the serial GERDQ score change did not exhibit any difference between the two groups. In the dexlansoprazole group, patients displayed persistent improvement in the GERDQ score in the on-demand period (week 8 vs week 24; P<0.001), whereas no continuous improvement was noted in the esomeprazole group (week 8 vs week 24; P=0.846)Table 3

Sequential changes of GERDQ scores during the 24-week study period

Esomeprazole is a potent PPI with single release characteristics that produces the maximum plasma concentrations at approximately 1.6-h post-dose. Daily administration of 40-mg esomeprazole yields good healing rates (87–94.1%) in patients with erosive esophagitis after 8 weeks.18,19,30 A meta-analysis revealed that 40-mg esomeprazole exhibited statistically significant improvement in erosive esophagitis compared with other PPIs such as 20-mg omeprazole, 30-mg lansoprazole, and 40-mg pantoprazole. A study reported a 5% (RR, 1.05; 95% CI: 1.02–1.08) relative increase in the probability of healing of erosive esophagitis with esomeprazole, yielding an absolute risk reduction of 4% and the number needed to treat of 25.31

Dexlansoprazole MR is the latest PPI that has been available in the United States for the treatment of acid-related disorders since 2009. The DDR formulation of dexlansoprazole extends the plasma concentration and the duration of acid suppression.16,17 Metz et al revealed that a daily dose of 60-mg dexlansoprazole MR controlled heartburn (median of 91–96% for 24-h heartburn-free days, 96–99% for heartburn-free nights) satisfactorily.16 In a clinical trial, the 8-week healing of erosive esophagitis was 92–95% using dexlansoprazole MR.17 Comparative trials of dexlansoprazole compared with 30-mg lansoprazole illustrated superior control in esophageal pH values, and the convenience of being able to dose the drug any time of the day regardless of the food intake.32

To the best our best knowledge, no direct head-to-head comparative study has investigated the long-term clinical effects or serial symptom scores between 60-mg dexlansoprazole and 40-mg esomeprazole. An indirect comparative study reported that 30-mg dexlansoprazole was more effective than 20- or 40-mg esomeprazole (RR, 2.01; 95% CI: 1.15–3.51; RR, 2.17; 95% CI: 1.39–3.38) in the symptom control of heartburn in patients with nonerosive reflux disease (NERD) after 4 weeks.33 In Japan, the cumulative healing rate of 20-mg esomeprazole during 24 weeks for GERD patients is 92.0% (88.0–96.0%).34 In a randomized, double-blind, placebo-controlled study, Johnson et al reported that the maintenance healing rate of erosive esophagitis was 93.6% after 24 weeks (95% CI: 87.4–99.7) in patients treated with 40-mg esomeprazole.35 In addition, Howden et al reported that 60-mg dexlansoprazole MR exhibited a significantly higher percentage of patients (86.6%) who maintained healed erosive esophagitis over 24 weeks compared with placebo (25.7%; P<0.00001) using the ITT population and life table analysis.36 Furthermore, Wu et al reported no significant differences between any doses of dexlansoprazole and esomeprazole in the maintenance of healed erosive esophagitis in the assessment of indirect comparisons.33

In this direct comparison study, the baseline acid regurgitation score was higher in the esomeprazole group than the dexlansoprazole group (3.3±0.6 vs 3.0±0.5; P=0.011). However, acid reflux sensation is one of the clinical symptoms of GERD, and heartburn score and epigastric acidity score were not statistically different between the two groups. In addition, the GERDQ score (contained 6 items) is a more objective questionnaire for the diagnosis and evaluation of the treatment efficacy20,43 of GERD than only one item. We observed no significant difference (23.2±3.7 vs 23.7±4.7; P=0.878) in the baseline GERDQ score of the two groups. Patients who completed the 8-week initial therapy were shifted to the on-demand treatment in the following study period. The overall CSR rates and the enhancement of the GERDQ score were similar between both groups. However, dexlansoprazole had lesser days with reflux symptoms in 24 weeks (53.9±54.2 vs 37.3±37.8; P=0.008). Moreover, we observed that patients exhibited persistent improvement in the GERDQ score in the on-demand period (week 8 vs week 24) in the dexlansoprazole group (P<0.001), but no continuous improvement in the esomeprazole group (week 8 vs week 24; P=0.846). Perhaps, different durations of drug retention exist in the circulation between two potent PPIs. According to a 1-week comparison study, 40-mg esomeprazole required more time (3 days) to attain CSR than 60 mg dexlansoprazole, especially in the female group because of estrogen and progesterone enhancing the relaxants of lower esophageal sphincter.37–39 In dexlansoprazole, no accumulation effect exists after multiple, once-daily doses of 60 mg, resulting in the maximum concentration (C_max) values of dexlansoprazole being slightly higher (<10%) on day 5 than on day 1.32,40 Consequently, dexlansoprazole could almost attain the target concentration on day 1. A one-day pH study comparing the pharmacokinetic effects of different PPIs after 12–24 h post-dose reported that the mean percentage of time with pH >4 and the average of mean pH were higher for dexlansoprazole than esomeprazole (60% vs 42%, P<0.001 and 4.5 vs 3.5, P<0.001, respectively).41 However, this study did not report the clinical effect after tablets were used. Fass et al reported that 84% of patients who were previously on twice‐daily esomeprazole were well controlled with once‐daily dexlansoprazole for the maintenance of heartburn symptom relief.42 This study noted a trend that there were a fewer number of tables of dexlansoprazole than esomeprazole (91.3±40.2 vs 96.7±44.9) and less GERDQ score at 16, 20, and 24 weeks in the on-demand treatment for the dexlansoprazole than esomeprazole groups (14.7±4.4 vs 16.2±4.7 (P=0.365), 13.7±3.2 vs 15.0±4.8 (P=0.124), and 13.1±3.8 vs 16.5±10.9 (P=0.252), respectively), although they were not statistically significant. Perhaps, it is because of too small volume of study cases to meet the significant change. Hence, dexlansoprazole could be a more optimal once-daily dose PPI for the on-demand use than esomeprazole.

This study has several limitations. First, this study managed to enroll a small number of patients and, therefore, did not attain the target number. Nonetheless, this preliminary pilot study is the first study to compare the clinical efficacy of a 24-week treatment with 60-mg dexlansoprazole and 40-mg esomerazole for LA grade A and B erosive esophagitis patients. Second, as the outcomes are subjective variables in this study, it could be essential to conduct a double-blind study to compare the treatment efficacy of two drugs.

In conclusion, this pilot study suggests that the symptom relief effect in 24 weeks for GERD was similar in dexlansoprazole and esomeprazole. Dexlansoprazole had lesser days with reflux symptoms in 24 weeks and better persistent improvement in the GERDQ score in the on-demand period. The dexlansoprazole could be a “more optimal” once-daily dose PPI for patients needing the on-demand use. Furthermore, these findings could have crucial implications for clinical practice in the on-demand period when treating patients with mild erosive esophagitis. As this issue was hampered by the small sample size in this study, comprehensive comparative studies are warranted in the future.