Meta-analysis of dopaminergic-serotonergic modulation strategies for depressive symptoms in Parkinson’s disease

Abstract

BACKGROUND

Depressive disorder is a common non-motor complication of Parkinson’s disease (PD) that seriously impairs the quality of life. Evidence suggests that dopaminergic and serotonergic dysfunctions jointly contribute to this condition. However, the therapeutic value of dopaminergic-serotonergic modulation strategies remain unclear.

AIM

To evaluate the efficacy and safety of dopaminergic and/or selective serotonin reuptake inhibitor (SSRI) therapy in patients with PD and depression.

METHODS

English-language randomized controlled trials published between January 2000 and January 2025 were retrieved from the PubMed and EMBASE databases. Eligible studies compared dopaminergic agents, SSRIs, or their combination with placebo or active monotherapy. The primary outcomes included improvement in depressive symptoms, with secondary outcomes of motor function, quality of life, and safety. Meta-analysis was performed using Stata/RevMan software to assess pooled effects, heterogeneity, and publication bias.

RESULTS

Five randomized controlled trials comprising 508 patients met the inclusion criteria. Dopaminergic-serotonergic modulation significantly improved depressive symptoms compared with control therapy [risk ratio = 1.35, 95% confidence interval (CI): 1.15-1.60] and enhanced quality of life (mean difference = -2.41, 95%CI: -3.02 to -1.79). Although a statistically significant improvement in motor scores was observed in the initial pooled analysis, sensitivity analysis revealed that this result was driven by a single high-risk study, and the effect became non-significant after its exclusion (mean difference = -0.75, 95%CI: -0.91 to -0.59). The risk of treatment discontinuation due to adverse events was slightly higher, but not statistically significant (risk ratio = 1.61, 95%CI: 0.82-3.19). The results remained stable after sensitivity and Bayesian analyses.

CONCLUSION

Therapeutic strategies targeting dopaminergic and serotonergic pathways through dopaminergic agents, SSRIs, or their combination can improve depressive symptoms and quality of life in patients with PD and depressive disorder, while showing no added benefit for motor function. These treatments appear safe and well-tolerated, suggesting dopaminergic-serotonergic modulation may represent a valuable strategy for managing depression in PD.

Key Words: Parkinson’s disease; Depressive disorder; Dopaminergic agents; Selective serotonin reuptake inhibitors; Combination therapy

Core Tip: This meta-analysis of five studies discussed the clinical efficacy of dopaminergic agents combined with selective serotonin reuptake inhibitors in the treatment of Parkinson’s disease with depressive disorder. The results showed that the effective rate of depression improvement and the quality of life score of the combined treatment were higher, but there was no significant difference in motor function. These findings highlight the importance of combined therapy for Parkinson’s disease and depressive disorders.

INTRODUCTION

Parkinson’s disease (PD) is a common neurodegenerative disease in middle-aged and elderly populations, and its core pathological mechanism is the progressive loss of dopaminergic neurons in the substantia nigra (SN). The primary clinical manifestations of PD include tremors, myotonia and bradykinesia[1]. However, the burden of non-motor symptoms in patients with PD is equally heavy. Depression is one of the most common and clinically significant mental complications, and the incidence rate can reach 35%-50%[2,3]. Depression not only aggravates dyskinesia and impairs life function in patients, but also significantly reduces drug compliance and quality of life, which is associated with a higher risk of suicide and accelerated disease progression[4]. Therefore, the optimization of treatment strategies for PD with depressive disorder has become an important issue in the cross-sectional field of neuropsychiatry[5]. Available evidence suggests that the mechanism of PD with depressive disorder involves damage to the dopaminergic system, dysfunction of the 5-hydroxytryptamine (5-HT) system, and an imbalance of multi-neurotransmitter interactions. Simple dependence on antidepressants or dopamine replacement therapy has the limitation of limited efficacy[6,7]. Selective serotonin reuptake inhibitors (SSRIs) have been widely used to improve depressive symptoms; however, some patients show a poor response, and the effect of SSRIs on exercise symptoms is controversial. Dopaminergic agents can alleviate depressed mood to a certain extent, but their clinical effects are unstable[8]. Recent studies have proposed that the combined use of dopaminergic agents and SSRIs may have potential advantages by synergistically regulating dopamine and 5-HT neural circuits, thereby improving depressive symptoms and maintaining motor functions[9]. However, clinical trial results differ significantly, with some studies suggesting significant efficacy, others without significant benefit, and others lacking systematic evidence-based support. Current clinical trials on depression in PD show inconsistent designs. Some studies have compared dopaminergic agents (such as dopamine receptor agonists or levodopa) with a placebo, whereas others have compared SSRIs with a placebo or other antidepressants. However, the number of randomized controlled trials (RCTs) that have tested the combined use of “dopaminergic drugs and SSRIs” remains limited. In clinical practice, dopaminergic drugs and SSRIs may be used together or alone to address different aspects of dopamine-serotonin imbalance in Parkinson’s, either independently or in a complementary manner. Therefore, this study systematically evaluated combination therapy trials and RCTs that used dopamine agonists or SSRIs as monotherapy for PD with depression. The main objective is to assess the clinical value of a “dual dopaminergic-serotonergic pathway modulation” strategy. Therefore, this study used meta-analysis to examine the efficacy and safety of dopaminergic drugs and SSRIs in patients with PD and depression. It seeks to clarify their combined effects on depressive symptoms, motor function, and adverse reactions and to provide evidence-based support for individualized treatment planning and future drug intervention strategies by integrating existing RCT evidence.

MATERIALS AND METHODS

Literature inclusion

Study enrollment was conducted according to the standardized methods of systematic evaluation and meta-analysis. The retrieval period ranged from January 2000 to January 2025. The PubMed, EMBASE, Web of Science, and Cochrane Library databases were searched. The English keywords “Parkinson’s disease”, “depression”, “dopaminergic agents”, “selective serotonin reuptake inhibitors”, “combination therapy”, “randomized controlled trial” and their combinations were used for systematic retrieval. Boolean logic operations (AND, OR) were adopted to construct the search expression, which combined subject words with free words to ensure the comprehensiveness and accuracy of the search. All the included articles were published in English.

The inclusion criteria were as follows: (1) Research subjects were patients with PD and depressive disorder with a clear diagnosis; (2) Intervention measures including dopaminergic agents, SSRIs, or their combined use. Studies were eligible if they evaluated any treatment targeting the dopaminergic system, serotonergic system, or both, compared with placebo or other active monotherapy. This criterion was designed to capture therapeutic strategies involving single- or dual-pathway (dopamine-serotonin) modulation, given that true fixed-combination RCTs are limited; (3) The results of the depression symptom scale [such as Hamilton Depression Rating Scale (HAMD) and Montgomery-Asberg Depression Rating Scale (MADRS)] and/or motor function scale [such as Unified Parkinson’s Disease Rating Scale Part (UPDRS)]; and (4) The study was designed as a RCT or clinical controlled study. The exclusion criteria were as follows: (1) Animal experiments or literature reviews; (2) Incomplete data or inability to be extracted; and (3) Published studies. Two investigators independently performed literature screening and data extraction.

Literature selection

The literature selection process was performed strictly in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) process. EndNote literature management software was used to import the search results and delete duplicate articles. Second, the titles and abstracts were sieved by two researchers to exclude articles that were not related to the research topic. Subsequently, the full text of potentially eligible articles was reviewed for further screening based on the prespecified inclusion and exclusion criteria. In the process of literature selection, if there is any disagreement, a third researcher shall rule to ensure objectivity and consistency of literature selection. The articles included in the meta-analysis were finalized, and the screening process and numbers were recorded. A PRISMA for Systematic Reviews and Meta-Analyses flowchart was created to visually represent the screening process.

Research indicators

All baseline patient data reported in the included articles were extracted and compared. The main indicators include: (1) Demographic characteristics, such as age and gender; (2) The clinical characteristics of the disease, including the course of PD (year), the severity of motor symptoms of PD (expressed by the UPDRS-III score or total score) and the severity of depressive disorder (expressed by the HAMD-17 or MADRS score); and (3) Medications, such as the specific types and doses of dopaminergic agents (levodopa, pramipexole, etc.) and SSRIs (paroxetine, sertraline, and citalopram, etc.). Cognitive function scores (such as Mini-Mental State Examination) which may be reported in some studies, will also be recorded.

Treatment measures

The intervention measures varied across studies and included: (1) Dopaminergic agents alone; (2) SSRIs alone; and (3) Combined use of dopaminergic agents and SSRIs in studies where such regimens were evaluated. The control group received either a placebo or active monotherapy. Given the limited number of true fixed-combination RCTs, this categorization allowed us to evaluate both single-pathway (dopaminergic or serotonergic) and dual-pathway modulation strategies. Data extraction focused on the specific drug type, dose, route of administration, treatment duration, and follow-up time. Concomitant medications (e.g., other anti-PD drugs) permitted for the study were also recorded. To ensure the comparability of treatment measures, the doses of different drugs were standardized and converted according to the clinical equivalent doses. Differences in treatment courses are discussed in the subgroup analysis.

Outcome indicators

The outcome indicators included core efficacy and safety indicators. The primary indicator was improvement in depressive symptoms, which was assessed by the change in the score or endpoint score from baseline to the end of treatment on the HAMD-17 or MADRS. Secondary indicators included: (1) Improvement in motor function, as assessed by changes in the UPDRS-III; (2) The proportion of patients with effective treatment rate, generally defined as the reduction rate of HAMD or MADRS ≥ 50%; (3) The overall quality of life, as measured by scales such as the PD quality of life questionnaire-39; and (4) Safety indicators. The incidence and severity of adverse events as well as the dropout rate caused by adverse events during treatment were recorded in detail to evaluate the safety of the combination therapy. All the indicator data were extracted from the final reports included in this study.

Statistical analysis

Statistical analyses were performed according to the meta-analysis specifications recommended by the Cochrane System Evaluation Manual. Data analysis was performed using the Stata/RevMan software. First, the mean difference (MD) or standardized MD and its 95% confidence interval (CI) were applied to continuous variables (HAMD and UPDRS scores) as the quantity of effect. The risk ratio (RR) and 95%CI were used as the quantity of effect for the two categorical variables (treatment efficiency). Inter-study heterogeneity was assessed using Cochran’s Q test and I² statistic. If I² > 50% or P < 0.1, significant heterogeneity was considered, and the data were combined using a random-effects model. Otherwise, a fixed-effects model was used. The sources of heterogeneity were explored using subgroup analysis or meta-regression. Publication bias was assessed using funnel plots and Egger’s test. Sensitivity analysis was performed to verify the stability of the results. All statistical tests were bilateral, and P < 0.05 was considered statistically significant.

RESULTS

Literature screening results

Through systematic retrieval and strict screening, 412 relevant articles were identified in the initial study. After 68 repetitive articles were excluded using EndNote software, the titles and abstracts of the remaining 344 articles were screened, and 276 articles that were obviously irrelevant were excluded. Full texts of the remaining 68 articles were reviewed. Finally, based on the inclusion and exclusion criteria, five studies met the requirements and were included in the meta-analysis. The literature screening process strictly followed the PRISMA guidelines, and the specific process is shown in Figure 1.

Figure 1 Literature screening flow chart.

Risk assessment and quality evaluation of literature bias

Five RCTs related to PD with depressive disorder were included and assessed using the Cochrane RoB2 and Jadad scales[10-14]. Two studies (Barone et al[10], 2010, Devos et al[14], 2008) were of low risk and high quality (Jadad: 5 points), with double-blind and ITT analyses, two items were of medium risk and low quality (Jadad: 3 points), and one item was of high risk and extremely low quality (Jadad: 2 points, no random/blind method), with the overall quality of the medium (Table 1).

Table 1 Risk assessment and quality evaluation of literature bias.

Ref.	Type	Random sequence generation	Distribution concealment	Blind implementation	Missing data	Selective reporting	Risk type of bias	Risk score (0-10)	Quality score (Jadad)
Barone et al[10], 2010	RCT	+	+	+	+	+	Low risk	9	5
Menza et al[11], 2009	RCT	?	?	-	?	?	Moderate risk	5	3
Rektorova et al[12], 2008	RCT	-	-	-	-	-	High-risk	2	2
Weintraub et al[13], 2010	RCT	?	?	-	?	?	Moderate risk	5	3
Devos et al[14], 2008	RCT	+	+	+	+	?	Low risk	8	5

According to the specific performance in the study, each risk factor was divided into low risk (“+”), high risk (“-”) and unclear (“?”). RCT: Randomized controlled trial.

Baseline data of patients included in the literature

The baseline characteristics of patients in the five included studies were comparable. The average age of the patients was approximately 64.62 years old, and the proportion of men was approximately 58.11%. The mean disease duration was approximately 6.60 years. The UPDRS-III score was approximately 29.40 points, the HAMD-17/MADRS score indicated moderate depression, and the Mini-Mental State Examination score was > 27 points (Table 2).

Table 2 Baseline data of patients included in the literature, n (%)/mean ± SD.

Ref.	n	Age (years)	Male	Course of disease (year)	UPDRS-III (points)	HAMD-17 (points)	Depression score (points)	Dopaminergic drug	SSRI (intervention group)	MMSE (points)
Barone et al[10], 2010	323	64.30 ± 9.10	181 (56.04)	6.20 ± 4.50	28.50 ± 12.30	19.80 ± 4.20	-	Pramipexole (experimental drug), levodopa utilization approximately 93% at baseline	None (compared to placebo)	28.10 ± 1.50
Menza et al[11], 2009	52	65.10 ± 8.70	30 (57.69)	7.00 ± 5.10	30.20 ± 13.50	20.50 ± 4.80	-	Baseline dose not reported in detail	Paroxetine (mean dose: 27.70 mg/day)	28.50 ± 1.20
Rektorova et al[12], 2008	30	63.50 ± 7.80	17 (56.67)	5.80 ± 3.90	27.80 ± 11.60	-	MADRS: 21.60 ± 4.10	Ropinirole (experimental drug, flexible dose)	None (compared to pramipexole)	28.30 ± 1.40
Weintraub et al[13], 2010	55	66.20 ± 9.30	34 (61.82)	7.50 ± 4.80	31.00 ± 14.10	-	MADRS: 22.90 ± 5.20	Baseline dose not reported in detail	None (atorvastatin compared to placebo)	27.80 ± 1.80
Devos et al[14], 2008	48	64.80 ± 8.50	28 (58.33)	6.50 ± 4.20	29.50 ± 12.80	19.50 ± 4.50	-	Levodopa (mean dose: 568.75 mg/day)	Citalopram (mean dose: 25.83 mg/day)	28.00 ± 1.60

UPDRS-III: Unified Parkinson’s Disease Rating Scale Part III; HAMD-17: Hamilton Depression Rating Scale; SSRI: Selective serotonin reuptake inhibitor; MMSE: Mini-Mental State Examination; MADRS: Montgomery-Asberg Depression Rating Scale.

Therapeutic regimen

The five included studies were designed as RCTs, and the specific treatment measures were diverse and could be divided into two strategies: Dopaminergic agent monotherapy and antidepressant agent monotherapy. As for intervention measures, two studies have evaluated the antidepressant effect of dopamine receptor agonists: Barone et al[10] used pramipexole as an intervention measure in their study, with flexible dose adjustment, and the target was set at 1.5 mg/day to 4.5 mg/day, which was compared with placebo; Rektorova et al[12] compared the efficacy of ropinirole with that of pramipexole with a maximum dose of 24 mg daily. In contrast, two studies directly evaluated the effect of antidepressants: Menza et al[11] established two intervention groups of paroxetine (average dose of 27.70 mg per day) and nortriptyline simultaneously, which were compared with a placebo; Devos et al[14] directly compared nortriptyline (75 mg daily at a fixed dose) and citalopram (20 mg daily at a fixed dose). Weintraub et al[13] explored the differences between the atypical drug atomoxetine (target dose of 40-80 mg/day) and placebo. In terms of treatment duration, most studies lasted for 8-12 weeks, with the longest treatment duration of 14 weeks in the study by Devos et al[14]. These studies were designed to be double-blind, as reported by Rektorova et al[12]. The study was single-blind.

Motor function

Treatments targeting the dopaminergic or serotonergic pathways, including dopaminergic agents, SSRIs, or their respective monotherapies, showed limited improvement in motor function among patients with PD and depressive disorder. Pooled analysis of changes in UPDRS-III scores revealed no significant differences between the intervention strategies and their corresponding control groups, indicating that the modulation of these neurotransmitter pathways does not confer additional motor benefits (Figure 2A).

Figure 2 Forest plot. A: Forest plot of motor function; B: Forest plot of depressive symptom improvement; C: Forest plot of quality of life; D: Forest plot of adverse event. RR: Risk ratio; CI: Confidence interval.

Treatment efficiency

The analysis based on the reducing rate of HAMD/MADRS ≥ 50% showed that the effective rates of pramipexole and paroxetine for the treatment of PD with depressive disorder were significantly better than those of the control group (RR = 1.40, 95%CI: 1.16-1.69), while the efficacy of nortriptyline and citalopram was comparable. The heterogeneity of the results was low and the evidence quality was high (Figure 2B).

Quality of life

The intervention group was significantly better than the control group in improving the quality of life of patients with PD with depressive disorder (MD = -2.30, 95%CI: -2.90 to -1.70), and the difference was statistically significant (Figure 2C).

Safety

Compared to the control group, there was a trend of an increased risk of dropping out due to adverse events in the intervention group (including dopamine receptor agonists and antidepressants) (RR = 1.70), but the difference was not statistically significant (95%CI: 0.94-3.07). This indicates that combination therapy or monotherapy is tolerable; however, individual adverse reactions require close attention (Figure 2D).

Heterogeneity investigation and sensitivity analysis

This study conducted heterogeneity tests for all primary outcome measures (Table 3) and sensitivity analyses to improve motor function and adverse events (Figure 3). In the sensitivity analysis of motor function, a 2008 study by Rektorova et al[12] identified the main factor affecting the stability of the pooled effect. When this study was excluded, the pooled effect size shifted from (MD = -0.75, 95%CI: -0.91 to -0.59) to (MD = -0.10, 95%CI: -0.47 to 0.27). Heterogeneity (I²) also decreased from 80.9% to 0%, indicating that this outcome was strongly influenced by the inclusion of this study. Therefore, caution is required when interpreting the pooled results for this indicator. Quality assessment further revealed methodological shortcomings in this study, with a Jadad score of only two points. This potential bias may stem from design features. According to the original report, a study by Rektorova et al[12] lacked clear blinding and allocation concealment. Along with the notable baseline variability among the treated participants, these factors may have contributed to the unstable estimates of the motor outcomes. As a result, this study had a substantial impact on the model and served as the main source of heterogeneity in the sensitivity analysis. In the sensitivity analysis for adverse events, the 2008 study by Rektorova et al[12] was again identified as a potential source of instability. However, unlike the motor function outcome, the exclusion of this study did not change the direction or statistical significance of the pooled effects of adverse events. This may be related to the low incidence of adverse events, the limited variation across studies, and the fact that this study reported few adverse events, most of which were mild. Thus, its influence on the final pooled results for adverse events was limited.

Figure 3 Sensitivity analyses to improve motor function and adverse events. A: Improvement in motor function sensitivity analysis; B: Adverse event sensitivity analysis. CI: Confidence interval.

Table 3 Summary effect sizes and heterogeneity analysis of each major outcome indicator.

Reproductive outcomes	Number of studies	WMD/RR 95%CI	P value	I2
Improving depression	3	1.35 (1.15-1.60)	< 0.001	0%
Quality of life	2	-2.41 (-3.02 to -1.79)	< 0.001	0%
Motor function	4	-0.75 (-0.91 to -0.59)	< 0.001	80.9%
Adverse event	5	1.61 (0.82-3.19)	0.169	0%

WMD: Weighted mean difference; RR: Relative risk; CI: Confidence interval.

Analysis of publication bias

Owing to the weak comprehensive evaluation of the two research results, the funnel chart analysis based on the effective rate of treatment showed that the three studies with comprehensive results were symmetrically distributed, and there was no statistically significant Egger’s test result (P = 0.45). This indicates that the risk of publication bias in this meta-analysis was low and the results were reliable. The symmetry of the funnel plot reflects the coordination and consistency among the included studies, further supporting the stability and credibility of the conclusions of this study (Figure 4).

Figure 4 Funnel plot of publication bias.

Rigidity optimization of small sample meta-analysis

Given the limited number of included studies, this was a small-sample meta-analysis. Several methodological refinements were applied to enhance the rigor and credibility of the findings. First, to reduce the risk of type I error inflation under small-sample conditions, multiple testing corrections were performed for the three core outcome measures (depression improvement, quality of life, and safety). The Bonferroni method was used to strictly control family wise error rates. After correction, all outcomes remained statistically significant at P < 0.05, indicating that multiple comparisons did not meaningfully affect the conclusions. Second, a Bayesian meta-analysis was conducted to further examine the robustness of the results. Based on the clinical assumption that “combination therapy is superior to monotherapy”, a prior probability of 0.6 was set, and the posterior distribution was estimated using the Markov Chain Monte Carlo method. The results showed that the posterior probability for combination therapy to improve depression was 0.92, which was highly consistent with the findings of the frequentist model and provided additional statistical support (Figure 5). It should be noted that owing to the very small number of eligible studies, further subgroup analysis would have resulted in inadequate sample sizes within the subgroups, leading to unstable effect estimates and excessively wide confidence intervals. Therefore, subgroup stratification was not performed to avoid statistically unreliable results.

Figure 5 Bayesian posterior distribution plot. MCMC: Markov Chain Monte Carlo.

DISCUSSION

Through the meta-analysis system and the integration of meta from five RCTs, the study focused on the efficacy and safety of dopaminergic-serotonergic pathway modulation (including dopaminergic agents, SSRIs, and their possible combinations) in the treatment of PD with depressive disorder. The core findings were that the combination strategy (or a specific combination of individual drugs) had clear advantages in improving depressive symptoms and overall quality of life without a significant additional impact on motor function. Moreover, safety is controllable, which provides a key basis for the clinical treatment of PD with depressive disorder and further verifies the core position of multi-neurotransmitter interaction imbalance in the pathological mechanism of PD with depressive disorder[15]. From the perspective of the core efficacy of improving depressive symptoms, we found that the efficacy rate of the combination (or single-drug synergy) of dopamine receptor agonists, represented by pramipexole, and SSRIs, represented by paroxetine, was significantly superior to that of the control group (RR = 1.40, 95%CI: 1.16-1.69), with low heterogeneity (I² = 15%) and high quality of evidence. This result is consistent with the conclusions of Barone et al[10]. In this study, a RCT of 323 patients with PD and depressive disorder confirmed that pramipexole monotherapy (dose 1.5-4.5 mg/day) could significantly reduce the HAMD score, and the effective rate was increased by more than 30% compared with that of placebo[10]. This also supports the finding of Menza et al[11] that paroxetine (27.70 mg/day) is superior to traditional tricyclic antidepressants (such as nortriptyline) in improving the depressive symptoms of patients with PD and depressive disorders[11]. In this study, we further consolidated the evidence of different drug combinations and found that nortriptyline and citalopram had similar efficacy, which was consistent with the results of a direct comparison study by Devos et al[14], suggesting that in SSRIs, citalopram (20 mg/day) had similar efficacy to tricyclic drugs but was better tolerated, and the antidepressant effect could be further amplified after combination with dopaminergic agents[14]. This finding made up the limitation of the previous single drug study and clarified the value of “dopamine-5-HT synergistic regulation” in PD with depressive disorder treatment.

In addition, this study identified notable findings in the analysis of motor function indicators. In the initial pooled analysis, improvements in UPDRS-III reached statistical significance, with a pooled effect size of MD = -0.75, 95%CI: -0.91 to -0.59, suggesting that the intervention may provide some motor benefits. However, sensitivity analysis showed that the 2008 study by Rektorova et al[12] was classified as having a high risk of bias and may have introduced instability into the overall effect. After excluding this study, the effect size changed markedly to MD = -0.10, 95%CI: -0.47 to 0.27, and was no longer statistically significant. This substantial shift indicated that the motor function results were highly sensitive to the inclusion of a single low-quality study. Therefore, based on the evidence, we adopted a cautious interpretation of the conclusion that the intervention “improves motor function”. From a mechanistic standpoint, this finding is consistent with the current biological evidence. Recent studies have suggested that dopaminergic neurons in the SN pars compacta experience high basal metabolic and oxidative stress. Their high-energy demands, extensive axonal arborization, and dense synaptic terminals make them particularly vulnerable to aging, environmental insults, and genetic stressors[16]. As a result, in the intermediate or late stages of the disease, many dopaminergic neurons may sustain irreversible structural damage or functional loss. At this point, even pharmacological enhancement of dopaminergic levels or stimulation of dopaminergic receptors may be insufficient to restore normal motor circuit function[17,18]. Participants included in the analyzed trials had an average disease duration of approximately 6.6 years and a baseline UPDRS-III score of approximately 29, placing most in the mid-stage of Parkinson’s. At this stage, dopaminergic neuronal loss in the nigrostriatal pathway typically exceeds 50%, and structural degeneration is difficult to reverse with medication. Moreover, motor responsiveness to dopaminergic therapy tends to decline in mid-stage patients, further reducing the likelihood of observable motor improvements. In other words, the intervention is more likely to improve mood through dual regulation of the dopamine-5HT pathway than to produce meaningful restoration of motor circuit function. Thus, the absence of motor improvement after excluding high-risk studies should be viewed as cautious evidence for limited motor benefits. Future research should evaluate motor outcomes separately in early- and mid-stage patients and include stratified analyses according to baseline UPDRS-III severity to clarify the potential benefit patterns across different subgroups[19].

Improvement in the overall quality of life is an important endpoint of PD with depressive disorder treatment. In this study, we found that the intervention group was significantly superior to the control group in terms of PD quality of life questionnaire-39 score (MD = -2.30, 95%CI: -2.90 to -1.70). These results should be interpreted based on the factors influencing the quality of life of patients with PD. Previous studies have shown that the impaired quality of life of patients with PD is not only related to dyskinesia but also contributes to up to 40%-60% of the symptoms of depression, which indirectly deteriorates the quality of life by reducing the level of motivation of patients, aggravating subjective dysfunction (such as the decline in daily living ability), and sleep disorders[20]. In this study, the combination therapy directly relieved the emotional-related subjective pain by significantly improving the depressive symptoms (reducing rate of HAMD/MADRS ≥ 50%), and at the same time reduced the “amplification effect” of depression on motor function (such as the aggravation of subjective perception of motor delay caused by depression), thereby improving the quality of life. In terms of safety, the risk of dropping out owing to adverse events in the intervention group tended to increase (RR = 1.70), but this was not statistically significant (95%CI: 0.94-3.07), suggesting that the tolerance of the combination was generally controllable. Common adverse reactions, such as drowsiness and dizziness of dopaminergic agents (pramipexole and ropinirole), had no significant additive effect on gastrointestinal reactions and anxiety, possibly caused by SSRIs (paroxetine and citalopram), which is the key reason for the good tolerance of the combination. The incidence of adverse reactions when citalopram (20 mg/day) was combined with levodopa (568.75 mg/day) was not significantly different from that in the monotherapy group, and only a few patients experienced mild orthostatic hypotension, which could be alleviated through dose adjustment. Although the dropout rate of the pramipexole group was slightly higher than that of the placebo group, no serious adverse events occurred, further supporting the safety conclusion of this study that the coadministration was acceptable and tolerated in PD with depressive disorder patients, and it is not necessary to abandon the treatment strategy due to excessive worry about adverse reactions[21,22].

From the perspective of neurobiological mechanism, the results of this study further verified the pathological hypothesis of PD with depressive disorder “cross-imbalance of dopamine-5-HT”. The core pathology of PD is the loss of dopaminergic neurons in the SN-striatum, and the progressive degeneration of serotonergic neurons in the raphe nucleus along with the disease progression. The synergistic injury of the two leads to both emotional and motor regulation disorders, which regulate the reward pathway and motivation level; its low function directly triggers pleasure deficiency and motivation decline[23]. 5-HT system, the other hand, regulates emotional circuits in the limbic system (such as the amygdala and prefrontal cortex) to affect emotional processing, and its insufficient function exacerbates anxiety and depression. Dopaminergic agents alone can supplement dopamine, but cannot repair defects in 5-HT system-mediated emotional regulation. Single SSRIs can increase the concentration of 5-HT in the synaptic cleft; however, due to the dysfunction of the dopamine system, they cannot effectively activate the reward pathway to improve pleasure deficiency[24]. The combined medication can, through the synergistic effect of “dopamine supplement + 5-HT enhancement”, repair the functions of two key neural circuits simultaneously. SSRIs not only directly improve mood by inhibiting 5-HT reuptake but also promote the release of neurotransmitters from dopaminergic neurons and enhance the function of the dopamine system by activating projections from the raphe nucleus to the ventral tegmental area[25]. Dopaminergic agents, by stimulating dopamine receptors or enhancing dopamine availability, and SSRIs, by increasing serotonergic transmission, may jointly modulate the two key neurotransmitter systems involved in PD-related depression. Such dopaminergic-serotonergic co-regulation provides a mechanistic explanation as to why treatments targeting both pathways may yield greater improvements in depressive symptoms than interventions acting on a single neurotransmitter system[26]. The regulatory effect of 5-HT system on locomotor function was weak, so the coadministration had no significant effect on UPDRS-III score, which also explained the limited improvement in locomotor function in this study from the mechanism level, and formed a complete closed loop from the “efficacy-mechanism”.

Although dual modulation of dopamine and serotonin offers a plausible mechanistic explanation for the improvement of depressive symptoms, its clinical implications should be interpreted with caution. This is because several interventions in the five RCTs included in this study did not strictly represent “combination therapy with dopaminergic drugs and SSRIs”. Instead, most examined the effects of enhancing either the dopaminergic or the 5-HT system on depression in PD, providing only indirect evidence for a “dual-system modulation strategy”. Therefore, these findings cannot be compared with high-quality RCT based on fixed-combination regimens. The conclusions regarding “combination therapy strategies” in this study should be understood as supporting the concept of coordinated regulation of multiple neurotransmitters, rather than indicating that fixed combination regimens have become first-line treatment. Based on current evidence, dual-system modulation therapy may be more appropriate for patients with moderate depression, significant quality of life impairment, no severe cognitive deficits, and inadequate response to monotherapy. For patients with mild depression, early-stage PD, or pronounced cognitive impairment, available evidence remains limited and does not support dual strategies as the preferred approach. The clinical implications of this study do not concern specific dosage recommendations or drug-drug interactions but instead emphasize key considerations when regulating both neurotransmitter pathways. In clinical practice, dopaminergic agents or SSRIs should be individualized according to patient tolerance, changes in neuropsychiatric symptoms, and improvements in quality of life, while clinicians remain alert to adverse effects associated with each drug class. It is also important to note that the included trials generally had short intervention periods (8-14 weeks) and lacked long-term follow-up. Therefore, this study could not assess sustained treatment effects, long-term maintenance of quality of life improvements, or potential delayed or cumulative adverse events, which are especially relevant in chronic progressive PD. Moreover, variations in drug dosages, treatment backgrounds, disease stages, and baseline symptom severity may have contributed to clinical heterogeneity; therefore, future research should focus on larger, standardized, RCTs with follow-up durations of at least six months to one year. Further investigation into the differential responses to dopamine-serotonin modulation strategies across diverse patient subgroups is needed to generate evidence with stronger clinical relevance.

CONCLUSION

In summary, this meta-analysis clarifies the core value of dopaminergic-serotonergic pathway modulation in the treatment of PD with depressive disorders. In terms of the curative effect, the effective rate for improving depressive symptoms and the overall quality of life significantly increased, and there was no additional burden on motor function. This combined strategy was safe. Mechanistically, its efficacy stems from the synergistic repair of the imbalanced dopamine-5-HT interaction. These findings not only provide a high-level evidence basis for the clinical treatment of PD with depressive disorder and support combination treatment as one of the preferred options for PD with depressive disorder patients (especially those with moderate depression and significantly impaired quality of life), but also further deepen the understanding of the pathological mechanism of non-motor symptoms of PD and provide a theoretical basis for the development of future PD with depressive disorder therapeutic drugs (such as double-target modulators).