Efficacy and safety of lansoprazole combined with flupentixol-melitracen for functional dyspepsia: A randomized, double-blinded, placebo-controlled clinical trial

Abstract

BACKGROUND

Psychosocial factors play a major role in the pathogenesis of functional dyspepsia (FD); however, there is a paucity of evidence from randomized controlled trials supporting the use of neuromodulators in FD.

AIM

To evaluate the efficacy and safety of lansoprazole (LPZ) combined with flupentixol-melitracen (FM) in FD.

METHODS

This was a single-center, double-blind, randomized, placebo-controlled trial enrolling patients aged 18-80 years who met the Rome IV diagnostic criteria for FD. Patients with a positive Helicobacter pylori test or abnormal upper gastrointestinal endoscopy/abdominal ultrasound were excluded. Participants completed the Leuven Postprandial Distress Scale, Short Form Nepean Dyspepsia Index (SF-NDI), Patient Health Questionnaire-9 (PHQ-9), Generalized Anxiety Disorders-7, and Pittsburgh Sleep Quality Index. Patients were randomized 1:1 to receive LPZ (30 mg once daily) combined with FM (flupentixol 0.5 mg + melitracen 10 mg) or matched placebo for 2 weeks, followed by a 4-week follow-up. The primary endpoint was clinical response rate in the intention-to-treat population, defined as a reduction of at least 0.7 in the postprandial discomfort syndrome (PDS) score at the end of treatment among patients with a baseline PDS score of ≥ 1.

RESULTS

Between March 5 and August 10, 2025, 183 patients were randomized to receive LPZ + FM (n = 92) or LPZ + placebo (n = 91). At week 2, the clinical response rate was higher in the LPZ + FM group than in the placebo group (63.0% vs 46.2%; relative risk = 1.42, 95% confidence interval: 1.04-1.93). LPZ + FM resulted in greater reductions in PDS, SF-NDI, PHQ-9, Generalized Anxiety Disorders-7, and Pittsburgh Sleep Quality Index scores. Improvements in PDS, SF-NDI, and PHQ-9 persisted through follow-up. Adverse events were mild and occurred in 32.6% of patients receiving LPZ + FM and 23.1% receiving placebo, with no serious adverse events reported.

CONCLUSION

Short-term, low-dose FM combined with LPZ is an effective and safe treatment for FD.

Key Words: Flupentixol-melitracen; Functional dyspepsia; Lansoprazole; Anxiety; Deanxit

Core Tip: This randomized, double-blind, placebo-controlled trial assessed the efficacy and safety of flupentixol-melitracen combined with lansoprazole for functional dyspepsia (FD). Our study demonstrated that this low-dose short-term regimen not only improved FD symptoms and quality of life within two weeks but also sustained efficacy for four weeks post-treatment. Future long-term follow-up and multicenter studies are warranted to verify its lasting benefits and health economic value, potentially advancing FD management from a “stepwise sequential” approach to a “brain-gut co-management” therapy.

INTRODUCTION

Functional dyspepsia (FD) is one of the most common disorders of gut-brain interaction, formerly known as functional gastrointestinal (GI) disorders[1]. According to the latest Rome IV criteria, FD is defined as the presence of symptoms originating in the stomach and duodenum, including early satiety, postprandial bloating, epigastric pain, and epigastric burning, in the absence of any organic, systemic, or metabolic disease that could explain these symptoms[2]. Based on these criteria, FD affects 2.2%-12.3% of the global population[3]. FD significantly impacts patients’ quality of life and imposes a substantial economic burden on healthcare systems[4]. Current guidelines recommend proton pump inhibitors, histamine-2-receptor antagonists, and prokinetics as first-line pharmacologic therapies for FD[5]. However, existing treatment regimens have several limitations. A Cochrane Database systematic review reported that proton pump inhibitors achieved an efficacy rate of only 31.1% in FD, indicating a limited therapeutic benefit of acid suppression[6]. In addition, safety concerns associated with prokinetics have restricted their clinical use[7]. Consequently, a substantial proportion of patients experience inadequate symptom control, repeated healthcare visits, and prolonged disease courses.

Notably, as a classic gut-related psychosomatic disorder, the onset and progression of FD are closely associated with psychological factors such as stress, cognition, and emotions[8]. Neuromodulators, particularly tricyclic antidepressants (TCAs), such as amitriptyline and imipramine, and certain antipsychotic drugs (such as sulpiride and levosulpiride), have been shown to effectively alleviate FD symptoms[9-11]. However, the clinical application of these drugs faces several challenges. First, they are typically introduced too late in the treatment course. Neuromodulators are usually reserved for refractory FD after failure of acid suppression therapy[5]. This sequential treatment strategy may trap many non-refractory FD patients in a “treatment-relapse” cycle, thereby delaying the optimal window for intervention. Second, neuromodulators generally have a slow onset of action, often requiring several weeks to achieve clinical efficacy, which may negatively affect patient adherence and early treatment confidence. In addition, drug adverse effects and tolerability are key concerns. Common side effects of TCAs, such as dry mouth and constipation, frequently lead to treatment discontinuation and directly compromise clinical efficacy[12,13]. Finally, limited awareness of neuromodulators among gastroenterologists further restricts their standardized use.

Flupentixol-melitracen (FM) is a fixed-dose combination of a TCA (10 mg melitracen) and a typical antipsychotic (0.5 mg flupentixol) that is widely used to treat mild to moderate depression and anxiety disorders[14,15]. Our previous clinical trial demonstrated that, compared with long-term (4 weeks) FM combined with omeprazole, short-term (2 weeks) FM combined with omeprazole reduced the risk of discontinuation syndrome while showing non-inferior efficacy in safely relieving symptoms of refractory FD[16]. Additional studies have reported that esomeprazole combined with FM significantly improved both GI symptoms and emotional status in patients with gastroesophageal reflux disease complicated by mood disorders, compared with acid suppression therapy alone without increasing adverse events[17]. Furthermore, a randomized crossover study suggested that short-term FM treatment improved quality of life in patients with FD, with efficacy independent of coexisting anxiety or depression. However, owing to sample size limitations, these findings require further validation[18].

Building on these preliminary data, we conducted a randomized controlled clinical trial to systematically evaluate the efficacy and safety of low-dose, short-term FM combined with lansoprazole (LPZ) vs LPZ monotherapy in patients with FD. We sought to determine whether early combined therapy with FM can achieve more effective and rapid symptom relief and improvement in quality of life, thereby providing evidence to optimize treatment strategies for FD.

MATERIALS AND METHODS

Study design and participants

This randomized, double-blind, placebo-controlled clinical trial recruited participants from the Gastroenterology Outpatient Department of Renji Hospital (Shanghai, China) between March 5, 2025, and August 10, 2025. Patients aged 18-80 years who met the Rome IV diagnostic criteria for FD underwent upper GI endoscopy and abdominal ultrasonography to exclude organic disease. Exclusion criteria were as follows: Organic lesions detected by upper GI endoscopy; positive Helicobacter pylori infection; severe dysfunction of vital organs (heart, liver, kidney, lung, etc.) or congenital disorders; pregnancy, lactation, or planned pregnancy; current or recent use (within 5 weeks) of monoamine oxidase inhibitors; known risk of narrow-angle glaucoma; or allergy to the study medications.

The trial consisted of a pre-randomization period of 7 days, a treatment period of 14 days, and a follow-up period of 28 days (Supplementary Figure 1). The study was conducted in accordance with the Declaration of Helsinki and the International Council for Harmonization Good Clinical Practice guidelines. The protocol was approved by the Clinical Research Ethics Committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong University, No. LY2024-323-C. Written informed consent was obtained from all participants prior to enrolment.

Randomization and masking

Eligible participants completed baseline assessments within one week before randomization. Patients were randomly assigned in a 1:1 ratio, using a computer-generated randomization list to receive LPZ (manufacturer: Takeda, Tianjin, China; brand: Takepron) combined with FM or placebo. Double-blinding was ensured by packaging FM (manufacturer: Haisco, Sichuan, China; batch no. 240751; brand: Lepan) and the matching placebo (manufacturer: Haisco, Sichuan, China; batch no. 240753) in identical sealed boxes with sequential numbering. The tablets were indistinguishable in packaging, shape, odor, and taste. The randomization codes were stored in opaque, double-sealed envelopes and maintained by designated personnel. Investigators, participants, and outcome assessors remained blinded to treatment allocation. Data collection and analysis were conducted under masked conditions. Before final unblinding, the safety and efficacy outcomes were evaluated by researchers who were unaware of group assignments. Participants were instructed to promptly present to the Renji Hospital Emergency Department if a treatment-related serious adverse event was suspected.

Procedures

Participants received either LPZ 30 mg plus FM (flupentixol 0.5 mg and melitracen 10 mg) or a matched placebo once daily in the morning for 2 weeks, followed by a 4-week follow-up period. Medication compliance was assessed by counting remaining tablets at the end of the 2-week treatment period. Adherence was considered good if at least 80% of the prescribed tablets for each medication were taken. Before treatment initiation, all participants received standardized education on FD and the study medications through dedicated outpatient counseling provided by experienced gastroenterologists. The core educational content emphasized the following points: (1) FD symptoms arise from abnormal gut-brain interaction rather than simple organic GI disease; (2) The neuromodulators used in this study relieve FD symptoms primarily by regulating the GI nervous system and are not psychiatric medications; and (3) Neuromodulators are administered using a low-dose, short-course regimen for FD, which differs from dosing strategies used in psychiatric practice. This standardized education aimed to improve patients’ understanding of FD pathophysiology and neuromodulator use, reduce stigma associated with antidepressant or antipsychotic treatment for GI disorders, and enhance informed consent and medication adherence[19].

The Leuven Postprandial Distress Scale is a validated daily diary consisting of eight items: Early satiety, postprandial fullness, upper abdominal bloating, epigastric pain, epigastric burning, nausea, belching, and heartburn. Each item is scored from 0 (none) to 4 (very severe)[20-22]. The postprandial discomfort syndrome (PDS) score is calculated as the mean of the early satiety, postprandial fullness, and upper abdominal bloating scores, while the epigastric pain syndrome (EPS) score is the mean of the epigastric pain and burning scores. Quality of life was assessed using the Short Form Nepean Dyspepsia Index (SF-NDI), which comprises 10 items, each scored from 1 (none) to 5 (very severe)[23,24]. Mood was assessed using the Patient Health Questionnaire-9 (PHQ-9) and the Generalized Anxiety Disorders-7 (GAD-7), consisting of 9 and 7 items, respectively, with each item scored from 0 (never) to 3 (almost every day). Sleep quality was evaluated using the Pittsburgh Sleep Quality Index (PSQI), which includes 7 components and yields a total score ranging from 0 to 21. Before randomization, all screened participants completed daily symptom diaries for one week. Patient-reported outcomes were collected at pre-randomization (T0, visit 0), baseline (T1, visit 1), end of treatment (T2, visit 2), and at 2 weeks (T3, visit 3) and 4 weeks (T4, visit 4) after treatment completion.

Outcomes

The primary endpoint was the clinical response rate. Clinical response was defined as a decrease of 0.7 or greater in the PDS score at the end of treatment (T2) among patients with a baseline PDS score of at least 1[22]. Minimal clinical response was defined as a decrease of 0.5 or greater in the PDS score at T2[20,25]. Scores were calculated as the mean value over the preceding week. Secondary endpoints included the minimal clinical response rate and changes in quality of life, mood, and sleep over time. Safety outcomes included the incidence of treatment-emergent adverse events (TEAEs) during the treatment period.

Statistical analysis

Based on prior studies, the efficacy rate of LPZ for FD was estimated at 50%[26]. Assuming an efficacy rate of 70% for LPZ combined with FM, an absolute efficacy difference of 20% was anticipated. Using a two-sided test with a significance level of α = 0.05 and 80% power, 91 evaluable participants were required per group, for a total of 182 participants. Allowing for an estimated 10% dropout rate, the planned sample size was 203 participants.

The full analysis set was analyzed according to the intention-to-treat (ITT) principle. The safety analysis set included all participants who received at least one dose of study medication and had post-treatment safety data. Data normality was assessed using the Kolmogorov-Smirnov test. Normally distributed variables are presented as mean ± SD, while non-normally distributed variables are presented as median (interquartile range). Missing data were imputed using the last-observation-carried-forward method. The per-protocol (PP) analysis set included all randomized participants who completed treatment as prescribed without major protocol deviations, including those who took at least 80% of the study medication and completed follow-up.

Between-group differences in efficacy rates were assessed using the χ² test, and relative risk (RR) with 95% confidence intervals (CIs) were calculated. Longitudinal changes in continuous outcomes, including PDS, EPS, SF-NDI, mood, and sleep scores at T1 through T4 were analyzed using linear mixed-effects models. Fixed effects included treatment group (LPZ + FM vs LPZ + placebo), visit time (T1 to T4), and the corresponding group by time interaction, with subject-specific random intercepts. Primary analysis focused on interaction effects and between-group changes from baseline. Bonferroni correction was applied to adjust for multiple comparisons across time points.

For prespecified subgroup analyses of clinical response rates, χ² tests were used to compare differences between subgroups, with RR with 95%CI calculated. All pre-specified variables (age, sex, duration of FD, baseline scores) were incorporated into exploratory logistic regression analyses. Compliance rates and the incidence of total TEAEs were compared using the χ² test. All statistical analyses were performed using SAS version 9.4, with a P value < 0.05 considered statistically significant. Figures were generated using GraphPad version 9.0.

RESULTS

Among the 204 patients with FD screened, 21 were ineligible, and 183 patients were enrolled and randomized. Of these, 92 patients were assigned to the LPZ + FM group and 91 to the LPZ + placebo group (Figure 1). Baseline demographic and clinical characteristics were well balanced between the two groups, with no significant differences in age, sex, duration of FD symptoms, dyspepsia subtype, symptom scores, quality of life, mood, or sleep quality scores (Table 1). A total of 21 participants withdrew or discontinued treatment: 12 in the LPZ + FM group and 9 in the LPZ + placebo group. Reasons for discontinuation included non-compliance with study medication in 11 participants (5 in the LPZ + FM group, 6 in the LPZ + placebo group), withdrawal of informed consent in 6 participants (4 in the LPZ + FM group, 2 in the LPZ + placebo group), and adverse events in 4 participants (3 in the LPZ + FM group, 1 in the LPZ + placebo group). There was no significant difference in treatment compliance between the two groups [94.6% (87/92) vs 93.4% (85/91), P = 0.742]. Overall, all 183 randomized patients were included in the ITT analysis. A total of 162 patients (80 receiving LPZ + FM, 82 receiving LPZ + placebo) completed the entire trial and were included in the PP analysis.

Figure 1 Trial profile. After excluding ineligible participants, 183 subjects were randomly assigned to the two treatment arms, and 162 completed the treatment course per protocol. 183 were included in the intention-to-treat analysis; 162 were included in the per-protocol analysis. PDS: Postprandial distress syndrome.

Table 1 Baseline characteristics, n (%)/median (interquartile range).

	LPZ + FM (n = 92)	LPZ + placebo (n = 91)	P value
Age, years	45.5 (12.2)	45.0 (12.3)	0.764
Body mass index, kg/m2	21.6 (2.7)	21.8 (2.6)	0.647
Sex			0.832
Female	55 (59.8)	53 (58.2)
Male	37 (40.2)	38 (41.8)
Education level			0.717
Primary school	35 (38.0)	37 (40.7)
High school or above	57 (62.0)	54 (59.3)
Duration of functional dyspepsia, months	12.0 (6.0-33.5)	13.0 (6.0-28.0)	0.730
Functional dyspepsia subtypes			0.628
Overlap	63 (68.5)	59 (64.8)
PDS	20 (21.7)	25 (27.5)
EPS	9 (9.8)	7 (7.7)
Clinical scores
LPDS-PDS	2.0 (1.3-3.0)	2.0 (1.3-2.7)	0.645
LPDS-EPS	1.0 (0.5-2.0)	1.0 (0.0-1.5)	0.165
SF-NDI	27.5 (19.3-34.0)	27 (20.0-34.0)	0.938
PHQ-9	7.0 (3.0-13.0)	8.0 (3.0-14.0)	0.939
GAD-7	5.0 (2.0-10.8)	5.0 (2.0-10.0)	0.948
PSQI	8.0 (5.0-12.0)	8.0 (5.0-12.0)	0.981

LPZ: Lansoprazole; FM: Flupentixol-melitracen; PDS: Postprandial distress syndrome; EPS: Epigastric pain syndrome; PHQ-9: Patient Health Questionnaire-9; GAD-7: Generalized Anxiety Disorder-7; PSQI: Pittsburgh Sleep Quality Index; LPDS: Leuven Postprandial Distress Scale; SF-NDI: Short Form Nepean Dyspepsia Index.

In the ITT analysis, patients with baseline PDS < 1 (9 in the LPZ + FM group, 7 in the LPZ + placebo group) were classified as non-responders. Among those receiving LPZ + FM, 63.0% (58/92) achieved a clinical response (decrease in PDS score ≥ 0.7) at T2, which was higher than that observed in the LPZ + placebo group [46.2% (42/91); RR = 1.42, 95%CI: 1.04-1.93; P = 0.022; Figure 2A]. When patients with a baseline PDS score < 1 were excluded, the clinical response rate in the LPZ + FM group [69.9% (58/83)] remained higher than that in the LPZ + placebo group [50.0% (42/84); RR = 1.56, 95%CI: 1.09-2.21; P = 0.009]. The PP analysis yielded similar results, with a higher overall FD clinical response rate in the LPZ + FM group compared with the LPZ + placebo group [65.0% (52/80) vs 47.6% (39/82); RR = 1.45, 95%CI: 1.03-2.03; P = 0.025; Supplementary Figure 2A].

Figure 2 Percentage of treatment response and changes of estimated Leuven Postprandial Distress Scale scores over time between lansoprazole plus flupentixol-melitracen and placebo group. A and B: Percentage of clinical response and minimal clinical response at the end of treatment. The relative risk, along with its 95% confidence intervals, was calculated to compare the lansoprazole + flupentixol-melitracen group against the lansoprazole + placebo group; C and D: Over-time model-based estimated postprandial distress syndrome score and epigastric pain syndrome score changes. The gray shaded area represents the treatment period. ^aP < 0.05, and ^bP < 0.01. RR: Relative risk; CI: Confidence interval; PDS: Postprandial distress syndrome; EPS: Epigastric pain syndrome.

The proportion of patients achieving a minimal clinical response (decrease in PDS score ≥ 0.5) was significantly higher in the LPZ + FM group than in the LPZ + placebo group [71.7% (66/92) vs 52.7% (48/91); RR = 1.54, 95%CI: 1.09-2.16, P = 0.008; Figure 2B]. In the completed case set, a significant between-group difference in the minimum clinical response was also observed [72.5% (58/80) vs 52.4% (43/82); RR = 1.59, 95%CI: 1.10-2.31; P = 0.008; Supplementary Figure 2B].

Changes in clinical scores from baseline over time were further evaluated. Significant between-group differences were observed in longitudinal changes in PDS and EPS scores throughout the study period (P < 0.0001 and P = 0.0002, respectively; Figure 2C and D). Compared with the LPZ + placebo group, the LPZ + FM group demonstrated a greater reduction in PDS score at T2 (week 2) [B = -0.38 (-0.64 to -0.11), P = 0.006]. No significant time by group interaction was observed for EPS scores at T2 [B = -0.13 (-0.33 to -0.06), P = 0.181]. However, at T3 (week 4), the LPZ + FM group showed a significantly greater reduction in EPS scores than the LPZ + placebo group [B = -0.19 (-0.38 to -0.00), P = 0.049]. In addition, after 2 weeks of treatment, the LPZ + FM group exhibited significantly greater reductions in SF-NDI, PHQ-9, GAD-7, and PSQI scores compared with the LPZ + placebo group (Table 2). Improvements in PDS, SF-NDI, and PHQ-9 scores persisted through T4 (end of trial) in the LPZ + FM group. In the completed case set analysis, EPS and PDS scores continued to show a significant time-by-group interaction (Supplementary Figure 2C and D), although reductions in EPS scores from baseline at individual time points were similar between the two groups (Supplementary Table 1).

Table 2 Model-estimated clinical outcomes over time between flupentixol-melitracen and placebo groups.

Variables	Time points	LPZ + FM	LPZ + placebo	B	95%CI	P value
LPDS-PDS	T1	2.0 (1.8-2.2)	1.9 (1.7-2.1)
	T2	1.0 (0.8-1.2)	1.4 (1.2-1.6)	-0.38	-0.64 to -0.11	0.006
	T3	1.0 (0.8-1.2)	1.3 (1.1-1.5)	-0.30	-0.57 to -0.04	0.026
	T4	1.1 (0.9-1.2)	1.4 (1.2-1.6)	-0.35	-0.61 to -0.09	0.009
LPDS-EPS	T1	1.1 (1.0-1.3)	1.0 (0.8-1.1)
	T2	0.5 (0.4-0.7)	0.7 (0.5-0.8)	-0.13	-0.33 to 0.06	0.181
	T3	0.5 (0.4-0.7)	0.7 (0.6-0.9)	-0.19	-0.38 to 0.00	0.049
	T4	0.6 (0.4-0.7)	0.7 (0.6-0.9)	-0.16	-0.36 to -0.04	0.109
SF-NDI	T1	27.3 (25.3-29.4)	27.4 (25.3-29.4)
	T2	15.7 (13.9-17.6)	18.8 (17.0-20.7)	-3.11	-5.74 to -0.48	0.021
	T3	14.9 (13.1-16.7)	18.2 (16.4-20.0)	-3.33	-5.89 to -0.77	0.011
	T4	15.5 (13.6-17.3)	18.8 (16.9-20.6)	-3.27	-5.89 to -0.65	0.015
PHQ-9	T1	8.5 (7.2-9.9)	8.6 (7.3-9.9)
	T2	3.2 (2.3-4.1)	5.3 (4.3-6.2)	-2.08	-3.36 to -0.80	0.002
	T3	3.1 (2.2-4.0)	4.8 (3.9-5.7)	-1.69	-2.93 to -0.46	0.008
	T4	3.2 (2.3-4.1)	5.3 (4.3-6.2)	-2.10	-3.38 to -0.82	0.001
GAD-7	T1	6.7 (5.5-7.9)	6.6 (5.4-7.8)
	T2	2.9 (2.0-3.7)	4.4 (3.6-5.3)	-1.55	-2.75 to -0.35	0.012
	T3	2.9 (2.1-3.7)	4.1 (3.2-4.9)	-1.16	-2.33 to 0.00	0.050
	T4	3.0 (2.1-3.8)	4.5 (3.6-5.3)	-1.48	-2.70 to -0.27	0.017
PSQI	T1	8.7 (7.8-9.6)	8.7 (7.7-9.6)
	T2	6.2 (5.3-7.0)	7.4 (6.6-8.3)	-1.28	-2.47 to -0.08	0.036
	T3	6.1 (5.3-6.9)	7.2 (6.3-8.0)	-1.07	-2.24 to 0.10	0.074
	T4	6.2 (5.3-7.0)	7.3 (6.5-8.1)	-1.15	-2.32 to 0.03	0.056

LPZ: Lansoprazole; FM: Flupentixol-melitracen; CI: Confidence interval; LPDS: Leuven Postprandial Distress Scale; PDS: Postprandial distress syndrome; EPS: Epigastric pain syndrome; PHQ-9: Patient Health Questionnaire-9; GAD-7: Generalized Anxiety Disorder-7; PSQI: Pittsburgh Sleep Quality Index; SF-NDI: Short Form Nepean Dyspepsia Index.

The pre-specified subgroup analysis based on baseline PDS score indicated that the treatment effect differed according to baseline PDS severity (interaction P value = 0.044). In contrast, no significant interactions were observed across subgroups defined by age, sex, education level, FD duration, baseline EPS score, or baseline PHQ-9, GAD-7, and PSQI scores (all interaction P > 0.05), suggesting that the treatment effect was generally consistent across these subgroups (Figure 3). Higher clinical response rates with LPZ plus FM were observed in the following specific subgroups: Patients with a baseline PDS score ≥ 2 (RR = 1.41, 95%CI: 1.03-1.93), a baseline EPS score < 1 (RR = 1.69, 95%CI: 1.14-2.50), an FD duration ≥ 1 year (RR = 1.48, 95%CI: 1.03-2.11), and those with a baseline PSQI score ≤ 10 (RR = 1.46, 95%CI: 1.01-2.09; Figure 3).

Figure 3 Predefined subgroup analyses for primary endpoint. Significance is given for interaction effect. LPZ: Lansoprazole; FM: Flupentixol-melitracen; RR: Relative risk; CI: Confidence interval; PDS: Postprandial distress syndrome; EPS: Epigastric pain syndrome; PHQ-9: Patient Health Questionnaire-9; GAD-7: Generalized Anxiety Disorder-7; PSQI: Pittsburgh Sleep Quality Index.

TEAEs were reported by 32.6% (30/92) of patients in the LPZ + FM group and 23.1% (21/91) in the LPZ + placebo group (P = 0.150). Drowsiness was the most common adverse event, occurring in 12.0% (11/92) of patients in the LPZ + FM group and 4.4% (4/91) in the LPZ + placebo group. Other adverse events included dry mouth [7.6% (7/92) vs 5.5% (5/91)], diarrhea [6.5% (6/92) vs 7.7% (7/91)], insomnia [4.3% (4/92) vs 2.2% (2/91)], nausea [4.3% (4/92) vs 3.3% (3/91)], and constipation [1.1% (1/92) vs 3.3% (3/91)] (Table 3). All adverse events were mild and resolved after treatment discontinuation.

Table 3 Summary of treatment-emergent adverse events, n (%).

TEAEs1	LPZ + FM (n = 92)	LPZ + placebo (n = 91)
Patients with any TEAEs	30 (32.6)	21 (23.1)
Drowsiness	11 (12.0)	4 (4.4)
Dry mouth	7 (7.6)	5 (5.5)
Diarrheal	6 (6.5)	7 (7.7)
Insomnia	4 (4.3)	2 (2.2)
Nausea	4 (4.3)	3 (3.3)
Constipation	1 (1.1)	3 (3.3)

¹A patient was counted only once, even if the same event was reported multiple times by the patient.

LPZ: Lansoprazole; FM: Flupentixol-melitracen; TEAEs: Treatment-emergent adverse events.

During the two weeks following treatment discontinuation, no cases of symptom recurrence were observed. Specifically, none of the patients experienced a return of PDS scores to baseline or higher levels at week 4 after achieving a minimal clinical response. Four patients newly reported anxiety during this period, including three in the LPZ + placebo group and one in the LPZ + FM group. All cases were mild (GAD-7 < 10), and no other withdrawal symptoms were reported.

DISCUSSION

This randomized, double-blind, placebo-controlled add-on trial demonstrated that a 2-week course of LPZ plus FM effectively reduced FD symptoms and improved quality of life. After two weeks of treatment, 63.0% of patients receiving LPZ plus FM achieved a clinical response, accompanied by a significant reduction in PDS scores. Patients in the LPZ + FM group also showed greater improvements in anxiety and depression symptoms, as well as sleep quality, at the end of treatment. Notably, these therapeutic benefits and quality-of-life improvements persisted through the end of follow-up. Although no significant between-group difference in EPS scores was observed at the 2-week time point, the overall reduction in EPS scores over time was significantly greater in the LPZ + FM group throughout the study period. The lack of significance at week 2 may be attributable to the limited room for EPS score reduction due to relatively mild baseline symptoms and insufficient statistical power related to the small sample size. Collectively, these findings support the potential of the LPZ + FM regimen as an initial treatment option for FD that provides rapid, sustained, and multidimensional symptom control.

As a chronic condition that substantially impairs quality of life, FD is associated with suboptimal treatment outcomes. Because some patients with FD exhibit heightened sensitivity to gastric acid compared with healthy individuals, acid suppression therapy and dietary interventions are often selected as initial treatments. However, these approaches provide only minimal symptom control[27-29]. Visceral hypersensitivity is a key pathophysiological mechanism underlying various disorders of gut-brain interactions, including FD. Similar to abdominal pain in patients with irritable bowel syndrome, many patients with FD experience discomfort or pain in response to normal gastric stimuli, such as gastric distension after food intake[30]. The complex bidirectional communication between the brain and gut plays a central role in FD pathogenesis and symptom persistence. Neuromodulators may improve FD symptoms by regulating the perception and processing of pain signals in the central and peripheral nervous systems. However, neuromodulators remain underutilized in routine clinical practice. Available evidence suggests that TCAs and antipsychotics, rather than selective serotonin reuptake inhibitors, provide therapeutic benefit in FD[31,32]. Amitriptyline and imipramine are preferred neuromodulators, with reported symptom improvement rates of 52.6% (51/97) and 63.6% (35/55), respectively[9,10]. However, the standard 12-week treatment duration presents challenges related to treatment adherence and tolerability, which may limit their clinical applicability. Completion rates in the amitriptyline and imipramine groups were 80% (78/97) and 73% (40/55), respectively. In contrast, the PP completion rate of the 2-week LPZ + FM regimen in the present study was 87% (80/92), with no statistically significant difference compared with the LPZ + placebo group (90%, 82/91). These findings suggest that the combination regimen does not impose an additional treatment burden. This approach not only addresses the limitation of delayed onset associated with conventional neuromodulators but also provides supportive evidence for its use as an initial treatment option in terms of therapeutic efficacy and practicality.

With respect to safety, adverse events were reported in both the LPZ + FM and the LPZ + placebo groups, none of which were severe. Drowsiness was the most frequently reported adverse event in the LPZ + FM group, consistent with findings from a clinical trial of FM in refractory chronic cough[33]. Melitracen, a TCA, may cause drowsiness through blockade of histamine H1 receptors, leading to somnolence. As a neuroleptic agent, flupentixol reduces central nervous system excitability by blocking dopamine receptors, particularly D2 receptors, which can result in sedation and lethargy[34]. Other commonly reported adverse events included dry mouth, diarrhea, insomnia, and nausea. However, all TEAEs observed in this study were mild and reversible. Notably, during the two weeks following treatment discontinuation, four cases of mild new-onset anxiety were reported (one in the LPZ + FM group), with no other withdrawal symptoms or symptom recurrence observed. Collectively, these findings support the favorable safety profile of this short-term treatment regimen. A retrospective study also demonstrated that low-dose antidepressant therapy for FD is both effective and well-tolerated[35]. Taken together, these data suggest that the combination regimen may be suitable for the initial management in a broad range of patients with FD, provided that patients are adequately informed and receive appropriate medication counseling.

To explore potential heterogeneity in treatment effects across patients with different clinical characteristics, a pre-specified subgroup analysis was conducted. An interaction between treatment effect and baseline PDS score was observed, suggesting that treatment efficacy may be influenced by the severity of baseline PDS symptoms. Concurrently, treatment efficacy was generally consistent across other prespecified subgroups, including demographics, disease duration, and baseline symptom scores, supporting the robustness of the primary findings across a broad FD population. Higher clinical response rates were observed in several specific subgroups, including patients with a baseline PDS score of ≥ 2, a baseline EPS score of < 1, and an FD duration of ≥ 1 year. Patients with higher baseline PDS scores (≥ 2) demonstrated more pronounced benefits. Clinically, patients with more severe symptoms are more likely to exhibit underlying pathophysiological abnormalities, such as impaired gastric accommodation or visceral hypersensitivity. As a neuromodulator, FM may exert therapeutic effects through dual mechanisms. First, by antagonizing dopamine D2 receptors, FM may enhance gastric motility and promote gastric emptying, thereby directly improving motility-related dysfunction in PDS. Second, its potential TCA component may modulate the brain-gut axis and attenuate visceral hypersensitivity. Although this study did not directly assess gastric emptying or visceral sensory thresholds, prior evidence supports these proposed mechanisms. Dopamine antagonists such as domperidone and metoclopramide have been shown to accelerate gastric emptying in patients with FD[36-38], while amitriptyline has demonstrated improvements in visceral sensitivity in experimental models and clinical studies[39,40]. These findings indirectly support the biological plausibility of FM acting through the aforementioned dual mechanisms. A higher clinical response rate was also observed in the subgroup with a lower baseline EPS score (< 1), although the interaction was not statistically significant, possibly owing to limited statistical power in this subgroup. Patients with an FD duration of ≥ 1 year also demonstrated higher clinical response rates, which may reflect chronic symptom persistence associated with long-standing central or peripheral sensitization, a key target of neuromodulatory therapy[41]. Patients with shorter disease duration may experience greater symptom fluctuation or spontaneous remission, potentially attenuating observed therapeutic effects. Overall, these findings suggest that baseline PDS severity may be an important determinant of treatment response. For patients with high baseline PDS scores, early combination therapy should be prioritized over conventional stepwise approaches. Findings from other subgroups should be interpreted cautiously.

Previous studies of antidepressant treatment for FD have largely focused on patients with refractory FD or comorbid mood disorders[10,42,43]. Notably, our findings showed no significant difference in the treatment effect of LPZ + FM between subgroups with mild-to-severe (baseline PHQ-9 or GAD-7 scores > 4) and the subgroup without mood disorders (score ≤ 4), indicating that the efficacy of LPZ + FM is independent of the presence of baseline mood disorders. These results suggest that the therapeutic benefits of FM may be mediated primarily through direct modulation of visceral sensitivity and brain-gut axis function, rather than through its anxiolytic or antidepressant effects. This interpretation is supported by an earlier crossover trial demonstrating the efficacy of FM in patients with FD who did not have diagnosed anxiety or depression[18]. Similar findings have also been reported for FM in the treatment of chronic cough[44]. In addition, a multicenter clinical trial reported a 53% remission rate with amitriptyline in non-depressed patients with FD[9]. Collectively, these data support the early integration of FM into FD treatment and highlight the need for future studies to further elucidate its direct regulatory mechanisms targeting GI-specific symptoms.

This study has several limitations. First, the follow-up period was relatively short, limited to 4 weeks after treatment completion, precluding assessment of long-term efficacy and symptom recurrence. Second, the primary efficacy endpoint relied on patient-reported symptom scores. Although this approach is standard in FD research, the lack of objective biomarkers may introduce recall bias. Third, as a single-center study with a limited number of patients with the EPS subtype, the robustness of subgroup analyses is constrained. Fourth, the study was conducted at a large tertiary center, enrolling patients with a median FD duration of one year and relatively complex clinical manifestations. This population likely represents a more refractory FD phenotype than that seen in primary care settings, introducing potential referral bias and limiting generalizability. Consequently, the observed therapeutic benefit may be more applicable to chronic or moderate-to-severe FD rather than newly diagnosed cases. Recognizing this limitation holds clinical relevance, as it can assist primary care physicians in determining when to escalate to this treatment regimen. In addition, the assumptions underlying the sample size calculation were somewhat idealized. Therefore, these findings should be confirmed in larger, more representative cohorts that include patients with early-stage or mild FD recruited from both tertiary and primary care settings.

CONCLUSION

To conclude, this study demonstrates that a low-dose, short-term treatment regimen of LPZ combined with FM provides meaningful clinical benefit in the management of FD. This combination achieved rapid, multidimensional symptom improvement and enhanced quality of life within two weeks, with effects maintained through follow-up. Subgroup analyses suggest that this strategy holds particular value for patients with moderate-to-severe PDS and a chronic disease course. Future long-term, multicenter studies are warranted to confirm the durability of these effects and to evaluate health economic outcomes, thereby supporting a shift in FD management from a traditional “stepwise sequential” approach toward an integrated “brain-gut co-management” approach.

ACKNOWLEDGEMENTS

The authors thank Bai-Song Wang, statistician at Renji Hospital, for technical support in statistics, and Sun Xia from the Department of Psychiatry for her assistance.