The effectiveness of using escitalopram in pediatric generalized anxiety disorder and the methods to predict the treatment response: A systematic review and meta-analysis

Introduction

Generalized Anxiety Disorder (GAD) is among the most prevalent mental health problems in children and adolescents, affecting an estimated 10%–15%.¹ It is characterized by excessive uncontrollable worrying and anxiety that could affect the affected child’s or adolescent’s relationships, situations, and day-to-day performance. Moreover, evidence shows a stronger association between anxiety disorders and suicidal ideation in these age groups than in the unaffected groups.² Despite all the aforementioned risks and the awareness of mental health that is constantly being raised among youth, a large proportion of young people with severe mental disorders have failed to seek and receive mental health care.³ Furthermore, little guidance in the diagnostic systems for the assessment of anxiety disorders to identify those in need of treatment makes the treatment of GAD challenging. In recent years, there has been a significant increase in the depth of understanding of the neurochemistry and functional neuroanatomy behind pediatric anxiety disorders, since data from functional neuroimaging suggest that pediatric patients with anxiety have functional abnormalities in the connections in the brain between the amygdala, medial prefrontal cortex, insula, ventrolateral prefrontal cortex, and dorsolateral prefrontal cortex.⁴ Selective serotonin reuptake inhibitors (SSRIs) are among the main treatment modalities used to treat pediatric anxiety, and are believed to improve the neurochemical and neuroanatomical changes that occur in GAD.⁵ Despite the availability of several other drug classes, SSRIs are considered the first-line management for pediatric anxiety.^5,6 Evidence to detect the efficacy and safety of SSRIs in pediatrics with a generalized anxiety disorder is still growing. Escitalopram is an SSRI often used to treat anxiety in adults. The exact mechanism of escitalopram is not fully understood, but it is believed that escitalopram increases serotonin levels in the brain by blocking presynaptic neuronal uptake. This leads to the improvement and relief of anxiety symptoms.⁷ However, the use, efficacy, and safety of escitalopram for pediatric anxiety are still under investigation. Additionally, the prediction of an escitalopram treatment response could significantly improve clinical intervention because it can improve patient outcomes by lowering the risk of adverse reactions as well as by eliminating the need for repeated iterations of trial-and-error testing, which minimizes costs and saves time.⁸ There is a shortage in the amount of data that helps clinicians assess which patients will respond to escitalopram treatment. In this study, we aimed to address the knowledge gap regarding the efficacy and safety of escitalopram in the treatment of pediatric generalized anxiety disorder and identify the potential predictors of escitalopram treatment response.

Methods

Results

Database search results

A PRISMA diagram (Figure 1) shows the details of the database search and screening processes. Following the retrieval of studies from the initial database search, 124 duplicate records were found and removed using EndNote X9 Software. The titles and abstracts of 305 records were screened against our eligibility criteria, yielding 55 records that warranted further full-text screening. Notably, the manual search did not yield any relevant studies. Fifty records were excluded; among these 50, three trials matched our eligibility criteria, but their results were not posted.^13–15 The reasons for exclusion are provided in Figure 1. Finally, five studies were included in this paper. Three studies were fit for quantitative and qualitative analyses, while the other two papers were analyzed qualitatively. For quantitative analysis, the main outcome was the PARS score, while for qualitative analysis, the main outcome was the side effects. Finally, a total of 5 studies were included in the qualitative and quantitative synthesis of our review.

Figure 1. PRISMA chart of the research flow.

Characteristics of the included studies

The outcome reporting bias was low risk in two of the included papers and high risk in three papers,¹⁶ which was evaluated as having a high risk of bias because the study did not report the effect of COVID-19 on anxiety and PARS scores in the results section despite mentioning it in the protocol, while in Refs. 16, 17 the change in γ-aminobutyric acid concentrations in the anterior cingulate was mentioned in the protocol and was not reported. The risk of bias (ROB) for all studies is illustrated in Figures 2 and 3. This study included 401 pediatric patients with GAD (a mean age of 13.5 years old). The baseline characteristics of patients are presented in Tables 2 and 3.

Figure 2. ROB for each included study.

Figure 3. Overall ROB of the included studies.

Table 2. Baseline characteristics of the included studies.

Study ID	year	groups	Sample size	Age (Mean ±SD)	Sex (Male %)	Conclusion
Strawn et al.17	2021	Escitalopram placebo	26 25	14.8±1.7 14.9±1.6	23% 24%	The first trail on the effect of escitalopram on pediatric patients with GAD. Escitalopram was superior to placebo in the treatment of pediatric GAD.
Lu et al.18	2021	Escitalopram placebo	21 20	14.9±1.7 15.0 ± 1.6	24% 30%	A trail to examine the effect of the escitalopram on the functional connectivity within Amygdala ventrolateral prefrontal cortex (VLPFC) circuitry and its role as a treatment predictor. the results showed that escitalopram significantly increased functional amygdala-VLPFC connectivity within 2 weeks, and this could predict the improvement in PARS score.
Lu et al.19	2022	Escitalopram placebo	19 17	14.5 ± 1.6 15.2 ± 1.5	4(24%) 6(30%)	In this study, the amygdala's functional connectivity was examined while the patient was in an emotional state. Correlation analysis was then conducted to see whether pretreatment and early (week 2) treatment-related connectivity were linked to treatment response. And it was discovered that escitalopram not only improved emotional processing but also increased the FC connection during emotional states.
Baumel et al.20	2022	Escitalopram placebo	based on the (Strawn et al. 2020) results			In this study, basal executive functions were examined as a therapy response predictor in adolescents with GAD, and it was discovered that escitalopram was most effective in young boys who had less severe working memory and emotional regulation issues. Yet, those who had a clinically significant range for organizing, planning, and job completion got better over time.
A Study of Escitalopram in the Treatment of Children and Adolescents With Generalized Anxiety Disorder16	2022	Escitalopram: placebo	45 91 44 93	9-11 11-17 9-11 11-17	31.1% 31.1%	Results showed that the effect ofescitalopram on pediatric patients with GAD was superior to placebo in the treatment of pediatric GAD

Table 3. Pre- and Post-treatment PARS scores.

Study ID	Baseline PARS score	PARS score 8 weeks	Follow-up duration n
Lu et al.18	17 ± 2 17 ± 3	7 ± 6 15 ± 3	8 weeks 8 weeks
Strawn et al.17	18±2 17±2 −8.65±1.31 compared to −3.52±1.06	9.35 13.48	8 weeks
Lu et al.19	17 ± 2 17 ± 3	12 ± 4 16 ± 3	2 weeks
Baumel et al.20	Based on Strawn et al. results
A Study of Escitalopram in the Treatment of Children and Adolescents With Generalized Anxiety Disorder)16	MD: -7.81(0.494)	MD: 6.38(0.484)	8 weeks

Escitalopram on PARS score: meta-analysis

The overall standardized mean difference between the escitalopram and placebo groups regarding the improvement in anxiety symptoms favored escitalopram (MD-4.53, 95% Confidence Interval (CI) [-7.75, -1.31, P=0.00001], and the pooled studies were heterogonous (Heterogeneity: Chi² = 2.83, I²= 99%) (Figure 4). To resolve heterogeneity, we conducted a sensitivity analysis in multiple scenarios, excluding one study in each scenario. Heterogeneity was best resolved by excluding AbbiVie (P=0.09, I²=65%). After removing the AbbiVie study from the meta-analysis model, the overall mean difference was still in favor of escitalopram over the placebo (Mean Difference (MD) -6.1, 95% CI [-8.75 to -3.44]; see (Figure 5)). According to the average baseline PARS score in the studies, the reduction was estimated to be nearly 35%.

Figure 4. Overall mean difference in PARS score before sensitivity analysis.

Figure 5. Overall mean difference in pars score after sensitivity analysis.

Discussion

The main aim of this study was to investigate the overall effect of escitalopram in minors (age: <18 years) with generalized anxiety disorder and to summarize the elements that can be used to predict the treatment response to escitalopram in these patients. The most obvious finding of this study was based on the overall mean difference of the included studies, suggesting that escitalopram is superior to placebo in the treatment of GAD in pediatric and adolescent patients. As it is known that escitalopram is recommended for treatment of major depressive disorder (MDD), generalized anxiety disorder, panic disorder, and social anxiety disorder (SAD) in adults, its use in minors is still under study. The first randomized control trial assessing the effect of escitalopram versus placebo was conducted by Strawn et al., which took into consideration the results of the previous meta-analysis on the use of other SSRI with GAD in youth.

The results of our study were similar to those of the study conducted by Isolan et al., in which all the patients showed significant improvement during escitalopram treatment.²¹

Two previous meta-analyses have been conducted on the effect of other SSRIs in pediatric anxiety, and (have concluded) the significance of the statistical and clinical improvement of anxiety symptoms; however, to the best of our knowledge, this is the first systematic review and meta-analysis that exclusively summarizes the effect and treatment response prediction of escitalopram in pediatric GAD.^22,23

Five eligible studies involving 401 patients with GAD were analyzed. The largest study yielded different outcomes; escitalopram’s advantage over the placebo was a marginal (-1) reduction in PARS, compared to the other studies (-5 in Ref. 17; -8 in Ref. 18), which accounts for the heterogeneity of the analysis. Unfortunately, this study did not publish the participants’ PARS scores, either at baseline or post-treatment with escitalopram, making it difficult to conduct a comparison and discover the cause of the heterogeneity. However, the results of this study indicate that escitalopram is superior to the placebo.

The PARS is built upon a checklist of 50 anxiety symptoms (covering SAD, SoP, and GAD) and 7 global items that are administered to the child and parent(s) together. Each of these global items is rated on a six-point (0–5) scale and reflects the number of presenting symptoms, their frequency, the severity of the feelings of anxiety, the severity of physical symptoms of anxiety, overall avoidance of anxiety-provoking situations, and anxiety-related interference with functioning at and outside of the home.²⁴ According to a study conducted by Caporino et al., a 35% reduction in the PARS score from the baseline is considered a treatment response, which is similar to the results of our study (average baseline was 17 and the MD was -6.²⁵ Similar results were found in a previous meta-analysis conducted by Baldwin et al. on the effect of escitalopram on social anxiety disorder and found that escitalopram was associated with a 33% mean decrease from the baseline.²⁶

Treatment prediction in patients receiving escitalopram was performed using different methods to predict treatment response and avoid the trial-and-error mechanism that is usually performed in the treatment of anxiety spectrum disorders. SSRIs treatment outcome prediction was performed to understand the variability in response to different drugs among different patients, and the results were described as clinically useful. In our paper, prediction for escitalopram was done by four different methods, according to Strawn et al., anxiety patients who were identified as intermediate metabolizers of CYP2C19 are predicted to have a greater response in treatment as opposed to rapid metabolizers, this is attributed to higher plasma concentrations of the drug in intermediate metabolizers. The CYP2C19 enzyme is the main metabolizer of escitalopram, thereby altering the drug activity according to its levels in different individuals, which depends mainly on the genetic factor, as the CYP2C19 gene has high polymorphism, with over 30 alleles.²⁷

Moreover, similar results were found in a study discussing the variability of CYP2C19 in treatment outcomes among children treated with escitalopram.⁶

In addition, results in adults were also similar; according to a retrospective study discussing the effect of CYP2C19 on treatment with SSRI in adults, intermediate metabolizers were associated with generally higher efficacy of SSRIs, and the study also showed that these patients were more susceptible to adverse effects of escitalopram treatment.²⁸

The second method reported is the role of functional neuroimaging (fMRI) in treatment prediction, and according to two papers conducted by Lu, et al., functional connectivity of the amygdala plays the main role. This could be explained by the fact that the amygdala has an intimate relationship with anxiety, as this brain center plays a crucial role in processing emotional stimuli such as fear and threat emotions. Most neurostructural studies on pediatric anxiety disorders suggest functional amygdala dysconnectivity as a biomarker for GAD.²⁹ A higher connectivity between the amygdala and right angular gyrus in patients receiving escitalopram was demonstrated by Lu et al., which serves as a factor for the prediction of improvement in anxiety symptoms at the 8th week. Functional brain activity in different brain regions was also found to be a treatment predictor for SSRIs in a study conducted by Preuss et al.; SSRI responders were observed to have a higher activity of the posterior cingulate gyrus, medial prefrontal cortex, and the thalamus during emotional response. Baseline amygdala-vmPFC connectivity and escitalopram-induced increase in amygdala-angular gyrus connectivity at week 2 predicted the magnitude of subsequent improvement in anxiety symptoms.³⁰

The use of amygdala connectivity has also been studied by Lu et al., but this time in conjunction with exposure to emotional stimuli. Functional MRI showed a change in functional connectivity between the left amygdala and the ventrolateral prefrontal cortex (VLPFC) from baseline to the 2nd week of treatment during emotional response, which predicted the improvement in treatment response in the 8th week, which could be useful in identifying escitalopram responders upon treatment initiation. The use of functional brain activity in different brain regions as a treatment predictor for SSRIs has been reported in several studies.^30–33

Furthermore, a study predicting escitalopram treatment response was performed by Harris et al., who explained that the change in functional connectivity between pretreatment and early treatment (second week) yielded significant results in the prediction of drug response.³⁴

This suggests that pretreatment and early treatment of amygdala connectivity could be used as an acute finding that helps predict which patients are more likely to respond to treatment as early as two weeks after starting the treatment course.

Additionally, it is worth mentioning that these changes described in the amygdala during emotional processing may show some contrast in anxious adults since a study reported that the magnitude of treatment response by venlafaxine was predicted by lower amygdala activity at the baseline.³⁵

The final method was the use of executive functions (EFs) as a predictor of treatment response, which is defined as a set of general-purpose control processes that are responsible for regulating behaviors and thoughts. It is linked to the prefrontal cortex, which is responsible for managing the dynamics of cognition and actions.³⁶ Assessment of the various EF subfields in pediatric patients with GAD at baseline can predict their response to treatment, as in the study conducted by Baumel et al. found that youths who had significant impairment in planning, organization, and task completion at baseline showed greater improvement, while those with significant baseline impairment in emotional control and working memory showed less improvement. The improvement in executive functions is considered very important and goes beyond treatment prediction, as executive function deficits in patients with generalized anxiety may exhibit poor inhibition, set-switching deficits, impairment of working memory, difficulties in planning, and task completion, all of which can affect their academic performance and interpersonal relations.^20,37 In addition, assessment of EF at baseline in anxious patients may be of great benefit for decision-making as some treatment options depend on certain executive functions; for example, anxiety and cognitive behavioral therapy CBT both require involvement of the executive functions as both require formation of future scenarios; anxiety formats a negative scenario while CBT formats a positive scenario, as such EF deficits may affect the CBT utility.³⁸ Figure 6 illustrates the take-home message from this paper.

Figure 6. Summary of the effects of the Escitalopram on the brain anatomy and neuroconnections.

Conclusion

Returning to the aim posed at the beginning of this study, escitalopram was shown to be effective in the treatment of GAD in pediatric and adolescent patients, and biomarkers such as functional connectivity, executive functions, and CYP2C19 enzyme activity were used to predict treatment response. This could help in decision-making and reduction of the repeated iterations of trial-and-error testing, which minimizes costs and saves time.