Oral Administration of East Asian Herbal Medicine for Inflammatory Skin Lesions in Plaque Psoriasis: A Systematic Review, Meta-Analysis, and Exploration of Core Herbal Materials

Psoriasis is an inflammatory autoimmune skin disease with various clinical manifestations. The aim of this review was to systematically evaluate the efficacy and safety of oral administration of East Asian herbal medicine (EAHM) for inflammatory skin lesions in psoriasis and to explore core herbal materials for drug discovery. A comprehensive search was conducted in 10 electronic databases for randomized controlled trials from their inception until 29 July 2021. Statistical analysis was performed in R version 4.1.2 and R studio. When heterogeneity in studies was detected, the cause was identified through sensitivity analysis, meta-regression, and subgroup analysis. Methodological quality was independently assessed using the revised tool for risk of bias in randomized trials. A total of 56 trials with 4966 psoriasis patients met the selection criteria. Meta-analysis favored EAHM monotherapy on Psoriasis Area Severity Index (PASI) 70 (RR: 1.2845; 95% CI: 1.906 to 1.3858, p < 0.0001), PASI 60 (RR: 1.1923; 95% CI: 1.1134 to 1.2769, p < 0.0001), continuous PASI score (MD: −2.3386, 95% CI: −3.3068 to −1.3704, p < 0.0001), IL-17, IL-23, TNF-α, and Dermatology Life Quality Index. Patients treated with EAHM monotherapy had significantly reduced adverse events incidence rate. In addition, based on additional examination of the herb data included in this meta-analysis, 16 core materials were identified. They are utilized in close proximity to one another, and all have anti-inflammatory properties. The findings in this study support that oral EAHM monotherapy may be beneficial for inflammatory skin lesions in psoriasis. Meanwhile, the identified core materials are expected to be utilized as useful drug candidate hypotheses through follow-up studies on individual pharmacological activities and synergistic effects.


Introduction
Psoriasis is an inflammatory autoimmune skin disease with various clinical manifestations, and there are millions of these patients worldwide [1]. The prevalence of this disease is reported differently in each country, and the overall prevalence is known to be between 0.14% and 1.99% [2]. Most patients with psoriasis are exposed to very negative psychological effects due to skin findings in exposed areas, such as the face and limbs, as well as shortened life expectancy due to complications of the disease [3,4]. The seriousness of the problem is also highlighted by the research findings, which show that more than 20% of psoriasis patients are depressed, which can lead to suicidal conduct in severe situations [4,5]. In addition, recent studies have reported that psoriasis is associated with various chronic diseases that can negatively affect life expectancies, such as psoriatic arthritis, hypertension, type 2 diabetes, dyslipidemia, myocardial infarction, and stroke [1,6,7]. This means that sions of various formulations and routes such as fumigation and ointments other than oral preparations are mixed. This suggests that it is difficult to see that the evidence for EAHM monotherapy with a specific route of administration has been established robustly. Therefore, at the present time, it is necessary to evaluate the efficacy and safety of EAHM for psoriasis based on a more rigorous study design for the route of administration and control group to be compared and to derive meaningful new drug candidate materials based on this data.
In accordance with the above recognition, we conducted a study according to the following objectives to provide clinicians with a clearer range of evidence, and at the same time, achieve the objective of exploring useful hypotheses for drug discovery: (1) efficacy and safety of EAHM monotherapy with the oral route of administration in inflammatory skin lesions of psoriasis are evaluated through the systematic review without limitation in scope. (2) Data mining on the herb data collected through this review is performed to derive a hypothesis related to the core EAHM material for psoriasis.

Methods
This study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis 2020 statement (Supplementary Table S1) [29]. The protocol of this systematic review was registered in PROSPERO (Registration Number: CRD42022296837, available from: https://www.crd.york.ac.uk/prospero/display_record. php?ID=CRD42022296837, accessed on 14 May 2022).

Search Strategy
Randomized controlled trials (RCT) that evaluated the efficacy and safety of EAHM monotherapy for plaque psoriasis were searched in the following 10 Supplementary Table S2.

Study Selection 2.2.1. Type of Studies
Only RCTs evaluating the efficacy and safety of oral administration of EAHM for plaque psoriasis were included. There were no restrictions on language and publication time. Some studies were excluded if they met the following criteria: (a) not RCT or quasi RCT; (b) not related plaque psoriasis or related disease; (c) primary intervention is not related EAHM; (d) not oral administration; (e) not clinical studies; (f) case reports or review; (g) not published in scientific peer-reviewed journals, including postgraduate theses or dissertations, and (h) when the experimental intervention is not EAHM monotherapy, such as combined therapy with conventional medicine.

Type of Participants
Trials were considered eligible for inclusion if they were conducted in patients with psoriasis, with no restriction on age, gender, or race. Since the subject of this review is plaque psoriasis, clinical trials that include patients with other subtypes of psoriasis such as psoriatic arthritis, guttate psoriasis, palmoplantar pulposus, and erythrodermic psoriasis were excluded from the review.

Type of Interventions
RCTs that compared EAHM as the active intervention in the treatment group versus placebo or CM in the control group were included. All forms of EAHM such as decoction, granule, capsule, compound preparation for the psoriasis treatment were included. There were no restrictions on the dose and duration of treatment for EAHM, but the mode of delivery was limited to oral intake. Studies in which East Asian medical interventions such as acupuncture, massage, or non-drug therapy were only combined in the treatment group were excluded. Studies in which the comparators included other EAHMs were excluded. Additionally, studies that were unable to verify the composition of specific herbal constituents that comprised the EAHM prescription utilized were omitted.

Type of Outcome Measures
The response rate of patients whose psoriasis area severity index improved by greater than 60% (PASI 60) and 70% (PASI 70), respectively, was employed as the primary endpoint. Meanwhile, the absolute difference between groups in PASI score was also used as the primary outcome. Secondary outcomes include tumor necrosis factor alpha (TNF-α), Dermatology Life Quality Index (DLQI), Interlukin-17 (IL-17), Interlukin-23 (IL-23). In addition, to evaluate the safety of the intervention for psoriasis patients, the incidence of adverse events (AEs) was also included as a secondary outcome.

Data Extraction
The titles and abstracts of potentially eligible studies were independently screened by 2 investigators (HGJ, HK) according to the above-mentioned search strategy. Afterward, a full-text review was performed based on the inclusion and exclusion criteria. Subsequently, information on the included studies was extracted independently by 2 reviewers (HGJ, HK). The following information was collected: title, author's name, clinical trial conducted country, diagnostic criteria, trial design publication year, sample size, participant age, sex distribution, interventions in the treatment and comparators, treatment duration, outcome index, reported adverse event, and composition with the dosage of EAHM. Any discrepancy was discussed with the third author (DL).

Methodological Quality Assessment
The methodological quality of each included study was evaluated independently by 2 investigators (HGJ, HK) according to the revised tool for risk of bias in randomized trials, Rob 2.0 [30]. It is comprised of five domains: bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in selection of the reported results. Methodological quality was assessed on three levels: "High risk of bias", "Low risk of bias" and "Some concerns". Disagreements between the two investigators were resolved with the help of the third author (DL).

Statistical Analysis Evidence Synthesis
Evidence synthesis of included studies with available data was performed by calculating the effect size and 95% CI using only the random effect model. Heterogeneity was considered statistically significant when the p-value based on the χ 2 test was less than 0.10 or I 2 was 50% or more. Two-sided p < 0.05 was considered statistically significant. Statistical synthesis of individual research results was performed in the software R version 4.1.2 and R studio program (Version 1.4.1106, Integrated Development for R. RStudio, PBC, Boston, MA, USA) using the default settings of the "meta" and "metafor" package [31]. The studies were grouped according to the type of intervention such as EAHM and comparator such as CM or placebo. Relative risk (RR) and 95% confidence interval (CI) were calculated for PASI 60 and PASI 70. Mean difference (MD) and 95% CIs were calculated for continuous PASI score and DLQI. For TNF-α, IL-17, and IL-23, standardized mean difference (SMD) and 95% CIs were calculated to integrate the results of several types of indicators for the same measurement target. Because the probability of an event that occurs was so much lower than other outcomes, and it was required to infer a causal relationship, AE was computed using odds ratio (OR). In this review, in order to effectively reveal the exact value of the effect size without relying only on the p < 0.05 significance threshold in the interpretation of the primary outcome synthesis result, a drapery plot was additionally illustrated along with the forest plot [32]. In the meta-analysis results, if heterogeneity was confirmed in an outcome that synthesized the results of more than 10 trials, the following additional analysis was performed to find out the cause. First, sensitivity analysis was performed according to the leave-one-out method to determine whether there was an effect by outliers in the included data. If no outliers are identified, after performing meta-regression analysis for the following three moderators specified in advance: (i) type of comparator, (ii) source of investigational medication, and (iii) type of EAHM formulation on the factors that had a substantial impact on the result, subgroup analyses were conducted. In order to distinguish publication bias, a contour-enhanced funnel plot was used for the outcome that included most studies [33]. For the asymmetry on the visually confirmed funnel plot, Egger's test [34] and Begg's test [35] were additionally performed to specifically confirm the existence of publication bias.

Hierarchical Agglomerative Clustering
The EAHM prescriptions used in each study reflect the medical goal of maximizing the synergy effect of the core herb combination. Therefore, hierarchical cluster analysis was used to understand the structure of the EAHM prescriptions used in individual studies. The analysis utilized in this study is agglomerative clustering, in which each observation is initially considered as a cluster of its own (leaf). Then, the most similar clusters are successfully merged until there is just one single big cluster (root).
The dissimilarity between individual herb constituents was considered as an individual distance, and the Euclidian distance was used as a measure of this. This corresponds to the shortest distance when it is assumed that the difference between each characteristic value is expressed on the coordinate plane.
Cluster analysis in this study was performed on herbal constituents that showed a frequency of occurrence of at least 20% compared to the total included clinical trials.

Social Network Analysis
To explore the interdependence of fundamental herbal constituents utilized in the EAHM prescription and to uncover the core material of connection, a social network analysis was performed on the herb data of individual studies in this review. On the surface, the "complexity" discussed in social network analysis looks to be perplexing, yet it is a term that suggests that an order based on the interrelationships of the constituent pieces exists. EAHM's prescription is an excellent illustration of the above-mentioned intricacy since it is guided by a combination of strict dosage principles and the tacit understanding of physicians who have worked with them for a long period. For this reason, the network analysis methodology has already been used in various ways in research analyzing EAHM [36,37]. Social network analysis in this review focused on two aspects. First, an undirected network was assumed, and the degree distribution was observed for the connectivity between the frequent herbal materials used in each EAHM prescription. In this case, since an undirected network is assumed, the average connection degree can be expressed as follows.
(n: number of nodes, E: number of links) Second, centrality was measured to identify herb materials with relatively large influence by comparing the influence of specific herbal medicines in the relationship between frequent herbs. Eigenvector centrality was used as the scale for the measurement that reflects the relationship between the individual herbs of EAHM that are prescribed at the same time. This scale can be expressed as: λ is the eigenvalue of herb i, a constant measured by the algorithm, and N(i) is the set of neighboring herbs of herb i. A ij becomes "1" if herb i and j have a connection in the n × n-direction adjacency matrix A, and "0" if there is no connection. In the case of C j , it is the eigenvector centrality value of herb j, which is herb i and neighboring herbs.

Quality of Evidence According to Outcome Measurements
The overall quality of evidence for each outcome was evaluated according to the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) pro [38]. The GRADE assessment evaluates the overall quality of evidence in four levels: very low, low, moderate, and high. The level of evidence is lowered according to factors, such as the risk of bias, inconsistency, indirectness, imprecision, and publication bias, respectively.

Study Identification
A total of 2434 studies were retrieved by electronic database search and manual search, among which 638 duplicate documents were removed. After screening the titles and abstracts, 1115 studies were excluded for at least one of the following reasons: (i) not related to psoriasis, (ii) primary intervention not related to EAHM, (iii) not oral administration, (iv) not clinical study (v) review article, (vi) case report or clinical experience, (vii) not a randomized controlled study. As a result of the evaluation of 460 articles for which full text was available among the remaining literature, 404 studies were excluded for the following reasons: (i) quasi-randomized controlled trials, (ii) duplicated documents, (iii) inappropriate study design, (iv) not disclosed herb ingredients, (v) not oral administration, (vi) not published peer-review scientific journal, (vii) not appropriate psoriasis subtype, (viii) not EAHM monotherapy, (ix) suspicion of salami slicing. Finally, 56 published studies were included in this review. Figure 1 shows the results of the database search. published studies were included in this review. Figure 1 shows the results of the database search.

Study Characteristics
The sample size of the included studies ranged from 40 to 260, and a total of 4966 participants were separated into the experimental group (n = 2605) and the control group (n = 2361). The psoriasis subtype in all included studies was psoriasis vulgaris or plaque psoriasis. One study was published in English, and all other studies were published in Chinese. The composition and formulation of the administered EAHM were reported in all studies included in this study. Only one study used a placebo preparation as a control group [39]; all other trials used CM as the control group. The following is a list of CMs that have been utilized as a control medication: methotrexate, vitamin A, glucocorticoids, and other topical medications including acitretin, compound amino-polypeptide agent, methotrexate, roxithromycin, penicillin, cephalosporin, vitamin A, glucocorticoids, and other topical agents. The duration of treatment in all eligible studies ranged from 2 weeks

Study Characteristics
The sample size of the included studies ranged from 40 to 260, and a total of 4966 participants were separated into the experimental group (n = 2605) and the control group (n = 2361). The psoriasis subtype in all included studies was psoriasis vulgaris or plaque psoriasis. One study was published in English, and all other studies were published in Chinese. The composition and formulation of the administered EAHM were reported in all studies included in this study. Only one study used a placebo preparation as a control group [39]; all other trials used CM as the control group. The following is a list of CMs that have been utilized as a control medication: methotrexate, vitamin A, glucocorticoids, and other topical medications including acitretin, compound amino-polypeptide agent, methotrexate, roxithromycin, penicillin, cephalosporin, vitamin A, glucocorticoids, and other topical agents. The duration of treatment in all eligible studies ranged from 2 weeks to 6 months. The characterization of the 56 included studies was summarized in detail in Table 1.

Risk of Bias
The methodological quality of 56 included studies was summarized in Table 2. The risk of bias in the studies was assessed by the Rob 2.0 tool [30]. The overall risk of bias in all studies was evaluated as "some concern". This is related to the fact that domain 2, domain 4, and domain 5 were evaluated as "some concern" in all studies except for one study [56]. All studies evaluated as "some concern" in domain 2 and domain 4 did not employ a double-blind design, and it is unclear whether the outcome assessor and the interventionist were clearly separated. In addition to this, it was not possible to confirm the pre-registered protocol in all studies. Due to this common problem, the risk of bias could not be completely excluded in all studies.   Figure 2A,B).

AEs
Among the included studies, 33 trials (34/56, 60.71%) reported information related to AEs [40][41][42]44,46,47,[49][50][51][54][55][56]58,59,[61][62][63]65,[67][68][69][70][71][72][74][75][76][77][78]80,84,87,92]. Four of these studies [46,50,58,77] did not report AEs in the control group, and two studies [62,65] reported the number of AEs in duplicate. On the other hand, there were five studies [51,63,67,84,92] that reported AEs in both groups. Therefore, 22 studies were able to synthesize the results by comparing the incidence rate. The aggregated results including 22 trials suggested that the incidences of AEs were significantly reduced by EAHM compared with CM control (22 trials, n = 2066; OR: 0.1017, 95% CI: 0.0630 to 0.1643, p < 0.0001; Figure 6). For the incidence rate of AEs, an additional comparison was performed through subgroup analysis according to the type of CM in the control group. Meta-analysis revealed that EAHM had lower incidence of AEs than amino-polypeptide agents (eight trials, n = 871; OR: 0.0939, 95% CI: 0.0399 to 0.2210, p < 0.0001; Figure 6). In comparison with acitretin, EAHM also showed a significant reduction in the incidence of AEs (10 trials, n = 976; OR: 0.0820, 95% CI: 0.0413 to 0.1628, p < 0.0001; Figure 6). Four studies comparing EAHM with other conventional medicines also showed a significant reduction in the incidence of AEs (four trials, n = 219; OR: 0.2428, 95% CI: 0.0879 to 0.6708, p < 0.0001; Figure 6). All the reported AEs were not severe and disappeared without long-term treatment. The details of adverse events reported in each study are recorded in Table 1. 3.6. Assessing Heterogeneity 3.6.1. Sensitivity Analysis Considerable heterogeneity was found in the synthesis of trial data using PASI 60 and continuous PASI score outcomes, with I 2 72% and 95%, respectively. In the drapery plot, there were also studies that appeared to be outliers. Accordingly, sensitivity analysis was performed according to the leave-one-out approach to determine whether a specific study corresponding to these outliers was the cause of heterogeneity for the above two results. As a result of the sensitivity analysis, as shown in Figure 7, each omission for all individual studies did not have a noteworthy effect on heterogeneity change ( Figure 7A,B).

Sensitivity Analysis
Considerable heterogeneity was found in the synthesis of trial data using PASI and continuous PASI score outcomes, with I 2 72% and 95%, respectively. In the drape plot, there were also studies that appeared to be outliers. Accordingly, sensitivity analy was performed according to the leave-one-out approach to determine whether a spec study corresponding to these outliers was the cause of heterogeneity for the above t results. As a result of the sensitivity analysis, as shown in Figure 7, each omission for individual studies did not have a noteworthy effect on heterogeneity change (Figure 7 B).

Meta-Regression and Subgroup Analysis
Through sensitivity analysis, it was confirmed that outliers in individual studies did not affect heterogeneity. Hence, in order to identify other potential causes of heterogeneity, a meta-regression analysis was performed on moderators expected to influence the results. The moderators to be evaluated were "type of comparator", "source of investigational medicine" and "sample size", and they were applied to the meta-analysis findings of PASI 60 outcome and continuous PASI score, respectively, and analysis was performed. As a result of performing a meta-regression for PASI 60, the type of comparator that was confirmed as a variable had a statistically significant effect on the pooled results (p = 0.0104; Figure 8), but the source of investigational medicine (p = 0.6945; Supplemetary Figure S1) and sample size (p = 0.8941; Supplemetary Figure S2) did not have a statistically significant effect. Neither the type of comparator (p = 0.1902; Supplemetary Figure S3), the source of experimental medicine (p = 0.5499; Supplemetary Figure S4), nor the sample size (p = 0.4478; Supplemetary Figure S5) had a significant influence on the effect size of studies in a metaregression of pooled results of continuous PASI score. Subgroup analysis indicated that the cause of heterogeneity may be related to the type of comparator (Table 3). Subgroup analysis was not performed for other predictors that were not significant in meta-regression. Meanwhile, for endpoints other than PASI 60 and continuous PASI score, additional sensitivity analysis, and subgroup analysis were not performed because the heterogeneity of the pooled results was low, or the number of included studies was very small. Through sensitivity analysis, it was confirmed that outliers in individual studies did not affect heterogeneity. Hence, in order to identify other potential causes of heterogeneity, a meta-regression analysis was performed on moderators expected to influence the Subgroup analysis was not performed for other predictors that were not significant in meta-regression. Meanwhile, for endpoints other than PASI 60 and continuous PASI score, additional sensitivity analysis, and subgroup analysis were not performed because the heterogeneity of the pooled results was low, or the number of included studies was very small.

Assessing Publication Bias
Contour-enhanced funnel plot, Egger's test, and Begg's test were used to assess the potential publication bias of the primary outcomes in this meta-analysis. Asymmetric

Assessing Publication Bias
Contour-enhanced funnel plot, Egger's test, and Begg's test were used to assess the potential publication bias of the primary outcomes in this meta-analysis. Asymmetric shapes were observed in the contour-enhanced funnel plots for all outcomes, suggesting potential bias ( Figure 9A-C = 0.1038). Overall, there may be a risk of potential publication bias, but it is difficult to say that such findings have been confirmed very clearly. Although no unequivocal evidence showing publication bias was found in the above investigation, the risk of potential publication bias could not be fully eliminated. PASI score was also significant in Egger's test, but no significant bias was confirmed in Begg's test (Egger's test: p = 0.0027; Begg's test: p = 0.1038). Overall, there may be a risk of potential publication bias, but it is difficult to say that such findings have been confirmed very clearly. Although no unequivocal evidence showing publication bias was found in the above investigation, the risk of potential publication bias could not be fully eliminated.

Quality of Evidence According to Outcome Measures
In the comparison between EAHM and CM, the overall quality of evidence according to all outcome measures was very low to moderate. The results of the GRADE assessment are presented in Table 4.

Quality of Evidence According to Outcome Measures
In the comparison between EAHM and CM, the overall quality of evidence according to all outcome measures was very low to moderate. The results of the GRADE assessment are presented in Table 4.

Detailed Information and Distribution of EAHM Ingredients
A total of 137 herbs were employed as component materials of the test EAHM in the 56 clinical trials covered in this review. Detailed information on individual EAHM components is summarized in Table 5. The following are 16 herbs that were prescribed with a high frequency in more than 20% of the studies included in this review: Rehmanniae Radix Recens; Salviae Miltiorrhizae Radix; Glycyrrhizae Radix et Rhizoma; Moutan Radicis Cortex; Lithospermi Radix; Smilacis Rhizoma; Radix Paeoniae Rubra; Dictamni Radicis Cortex; Imperatae Rhizoma; Hedyotidis Herba; Isatidis Radix; Lonicerae Flos; Sophorae Flos; Scutellariae Radix; Forsythiae Fructus; Spatholobi Caulis. The relative frequencies of these top 16 herbal materials ranged from 21.43% to a maximum of 69.64%. In terms of herb properties, all thirteen herbs, with the exception of three, were classed as cold and had the highest proportion, two herbs were neutral, and one herb had a warm property. Herbal flavors could be classed as bitter or sweet; however, bitter herbs accounted for a bigger part of the total, with nine herbs. Hence, the specific efficacy that clinicians consider when prescribing EAHM is expressed as summary information called the "action category".    Anemarrhenae Rhizoma 10 g, Rehmanniae Radix Recens 10 g, Glehniae Radix 10 g, Liriopis seu Ophiopogonis Tuber 10 g, Gardeniae Fructus 10 g, Lilii Bulbus 10 g, Lophatheri Herba 10 g, Zizyphi Semen 10 g, Phellodendri Cortex 10 g, Corni Fructus 9 g, Prunellae Spica 10 g  Table 5. Cont.

EAHM Prescription Name Source Ingredients of EAHM Prescription (Latin Name) Types of Preparation
Wen ( The action categories of the 16 high-frequency herbs mentioned above were all classified as "heat-clearing" except for 1. Table 6 shows the classification information for 16 herbs, including frequency distribution, property, taste, and action category.

Hierarchical Agglomerative Clustering
The characters of the top 16 high-frequency herbal materials were investigated using the hierarchical agglomerative cluster method. Through this analysis, pharmacological trends of core EAHMs used in the treatment of inflammatory skin lesions in psoriasis can be identified. The core herbs in this study may be separated into three modules as a result of classification based on the frequency of use and features of individual herbs. The results of classifying herbs are shown in Figure 10. The characters of the top 16 high-frequency herbal materials were investigated using the hierarchical agglomerative cluster method. Through this analysis, pharmacological trends of core EAHMs used in the treatment of inflammatory skin lesions in psoriasis can be identified. The core herbs in this study may be separated into three modules as a result of classification based on the frequency of use and features of individual herbs. The results of classifying herbs are shown in Figure 10.

Social Network Analysis
Social network analysis was used to confirm the mutual relationship between 16 herbs used frequently for inflammatory skin lesions of psoriasis and to identify core materials showing higher centrality in this interrelationship. As a result of graphically expressing the network between each herb, it was found that they are all closely connected, as shown in Figure 11. In the calculation of eigenvector centrality to measure the prestige centrality of individual herbs, Sophorae Flos and Scutellariae Radix were 0.0593, and all other 14 herbal materials were 0.0630. According to this, the centrality of 16 high-frequency herbs used in more than 20% of trials was generally at a similar level, and it could be interpreted that they were considered closely related to each other in their use in EAHM prescription for psoriasis.

Social Network Analysis
Social network analysis was used to confirm the mutual relationship between 16 herbs used frequently for inflammatory skin lesions of psoriasis and to identify core materials showing higher centrality in this interrelationship. As a result of graphically expressing the network between each herb, it was found that they are all closely connected, as shown in Figure 11. In the calculation of eigenvector centrality to measure the prestige centrality of individual herbs, Sophorae Flos and Scutellariae Radix were 0.0593, and all other 14 herbal materials were 0.0630. According to this, the centrality of 16 high-frequency herbs used in more than 20% of trials was generally at a similar level, and it could be interpreted that they were considered closely related to each other in their use in EAHM prescription for psoriasis.
Nutrients 2022, 14, x FOR PEER REVIEW 41 of 52 Figure 11. Core EAHM material network for treating psoriasis. The weight reflects the frequency at which two herbs are paired and utilized simultaneously in one EAHM prescription.

Summary of the Main Finding
Through the above analysis, our meta-analysis results suggest that oral EAHM is effective in improving symptoms of psoriasis. Overall, in the clinical trials included in this study, EAHM as monotherapy showed superior skin manifestation improvement in psoriasis compared to placebo and CM active controls in PASI 60, PASI 70, and continuous PASI indexes. At the same time, EAHM showed a superior or similar level of an effect to CM on the inflammatory findings of psoriasis in indicators such as IL-17, IL-23, and TNFα, and also showed positive results on the quality of life in psoriasis. On top of that, patients treated that EAHM were more likely to experience less incidence rate of AEs. In this review, 16 high-frequency materials were derived through separate data mining of the collected herbal prescription information. Most of these herbs showed a clear tendency of property cold and action category "heat-clearing", and it was found that all herbal materials were used with close correlation within the EAHM prescription.

Strength and Implications of Clinical Practice
The strength of this study is that we focused on the efficacy and safety of EAHM by the oral route of administration and as monotherapy alone. Since the efficacy that can be confirmed through clinical studies on combined therapy is an add-on effect, it should be viewed as essentially different from the efficacy of monotherapy of the intervention. On the other hand, even for materials with the same pharmacological effect, the fact that pharmacokinetics will vary depending on the administration route is no exception for natural products [95,96]. Recently, as the scope of research on pharmaceuticals based on natural sources continues to expand, the development of inhalation aerosol or injections is being Figure 11. Core EAHM material network for treating psoriasis. The weight reflects the frequency at which two herbs are paired and utilized simultaneously in one EAHM prescription.

Summary of the Main Finding
Through the above analysis, our meta-analysis results suggest that oral EAHM is effective in improving symptoms of psoriasis. Overall, in the clinical trials included in this study, EAHM as monotherapy showed superior skin manifestation improvement in psoriasis compared to placebo and CM active controls in PASI 60, PASI 70, and continuous PASI indexes. At the same time, EAHM showed a superior or similar level of an effect to CM on the inflammatory findings of psoriasis in indicators such as IL-17, IL-23, and TNF-α, and also showed positive results on the quality of life in psoriasis. On top of that, patients treated that EAHM were more likely to experience less incidence rate of AEs. In this review, 16 high-frequency materials were derived through separate data mining of the collected herbal prescription information. Most of these herbs showed a clear tendency of property cold and action category "heat-clearing", and it was found that all herbal materials were used with close correlation within the EAHM prescription.

Strength and Implications of Clinical Practice
The strength of this study is that we focused on the efficacy and safety of EAHM by the oral route of administration and as monotherapy alone. Since the efficacy that can be confirmed through clinical studies on combined therapy is an add-on effect, it should be viewed as essentially different from the efficacy of monotherapy of the intervention. On the other hand, even for materials with the same pharmacological effect, the fact that pharmacokinetics will vary depending on the administration route is no exception for natural products [95,96]. Recently, as the scope of research on pharmaceuticals based on natural sources continues to expand, the development of inhalation aerosol or injections is being actively carried out depending on the disease, as well as being used as external preparations such as ointment or fumigation [97][98][99]. Therefore, in the design of future EAHM studies, a clear definition of the administration route is bound to be a more important requirement. This study was aimed at deriving hypotheses related to candidate materials and indications for oral drugs beyond a simple meta-analysis, and there is no dispute that the route of administration and the conditions of monotherapy are important.
Evidence in this study derived according to the above scope suggests that oral administration of EAHM monotherapy is a useful option for inflammatory skin lesions management in psoriasis. The primary finding of this study is that the response rate and severity of PASI can be significantly improved. In addition, the improvement effect of various inflammation-related outcomes and DLQI is also a valuable finding in this study. These results are more meaningful in that they are consistent with several previous reports [21,23,100]. Therefore, administration of EAHM may be attempted as an indication for skin damage accompanied by inflammation in psoriasis patients. It seems reasonable to use EAHM for patients who show low compliance or do not respond to conventional CM treatment. Another important finding to consider is that when EAHM is used, the incidence of AEs is significantly reduced. Despite the need for systemic treatment through oral agents, it is worthwhile to apply EAHM monotherapy as an alternative to patients whose side effects of CM are too pronounced. Further analysis of the EAHM prescription data revealed that herb materials with specific properties were used frequently for psoriasis. Accordingly, the commonly prescribed core herbal material of this review and their close interrelationships information can help in the selection and combination of the appropriate herb when constructing customized EAHM formulations for individual patients.

Implications of Core Material Exploration
For the effective indications of EAHM for psoriasis revealed in the above discussion to be linked to the development of new drugs, further exploration of mechanisms and key materials is required. In this process, two important characteristics of EAHM must be considered first. One of them is related to the diagnostic method of East Asian medicine, which separately classifies the tendency to show systemic syndromes in addition to the patient's biomedical symptoms and pathology [101]. Such a diagnostic method that can administer customized prescriptions for the same disease is called "pattern identification" or "syndrome differentiation". The properties and action categories assigned to individual EAHM herbs represent therapeutic targets according to this diagnosis [27,[102][103][104]. Specific EAHM indications have been primarily differentiated using the notions of "cold syndrome" and "hot syndrome," and medications with "hot property/cold property" have been administered in response. For example, when a patient diagnosed with psoriasis complains of inflammatory skin symptoms along with physical findings such as fever, sweating, and thirst, it can be subdivided into hot syndrome of a psoriasis patient. EAHM materials that can effectively alleviate the accompanying systemic findings of this type of 'hot syndrome' are classified as cold properties. Conversely, EAHMs that can control cold syndrome are generally classified as hot property [105]. Recent studies exploring this topic at the molecular mechanism level have shown that EAHM, classified as a hot property, is implicated in pathways that include neurotransmitter reuptake, cold-induced thermogenesis, blood pressure regulation, and adrenergic receptor signaling. In the case of cold property, there are reports that the target gene is related to the steroid pathway. As a consequence, the hot/cold properties of EAHM were presumed to be the major factors in this study, implying distinct signals and mechanisms of action, which were incorporated in the analysis [106].
Most of the 16 high-frequency core herbs identified in this study were materials that exerted "clearing heat" action based on the "cold" property, and cluster analysis also confirmed that many herbs can be clustered with similar properties. This implies more information than simply that "clearing heat herb" is frequently used to manage inflammatory skin symptoms of psoriasis. According to previous studies, herbs exhibiting "clearing heat" action among EAHM are known to exhibit various anti-inflammatory and antiviral effects on patients with the so-called "heat" symptom pattern [107]. Hence, a more recent study revealed that "medicinal herbs of clearing heat" had multiple anti-inflammatory activities compared to herbs belonging to other action categories [108]. As summarized in Table 7, the pharmacological activity of the core herbs in this study is consistent with the knowledge in previous studies in that they correspond to anti-inflammatory and immune-modulating actions by various pathways. Therefore, the clinical efficacy of EAHM on psoriasis observed in this review appears to be strongly related to the complex antiinflammatory mechanism exerted by herbs belonging to the "clearing heat" category. At the same time, in the future EAHM drug discovery related to psoriasis, it is expected that drugs corresponding to the above-discussed categories can be considered as preferred candidate materials.  Kim (2007) [113] Lithospermi Radix -rat peritoneal mast cells -PCA rat Shikonin -Inhibiting the release of histamine in a dose-dependent manner -Inhibiting the anti-DNP IgE-induced passive cutaneous anaphylaxis reaction and IL-6, IL-8, and TNF-α expression -Inhibiting NF-κB activation and IκB-alpha degradation  On the other hand, another characteristic of EAHM that should be considered is the synergistic effect exerted through multi-compound action against the multi-target [28,125,126]. As can be seen from the data in this review, EAHM is usually administered in the form of a polyherbal formulation. This formulating chemical compound of EAHM not only produces a better synergistic effect, but also exerts an effect on the complex underlying mechanism of various diseases by reducing the side effects of individual drugs [18,127,128]. The main prescription principle of EAHM that makes this possible is expressed as "Gun-Shin-Jwa-Sa" (King-Retainer-Officer-Messenger in English words) [18,24]. In this approach, herbs responsible for the main effect are placed in a higher dose ratio at the positions of "Gun" and "Shin", while herbs that lessen medication side effects or boost synergy are placed in relatively small doses at the positions of "Jwa" and "Sa". Through this, an appropriately composed herbal combination can be expected to have amplified efficacy compared to that of a single herb. For example, the EAHM formula composed of only Sophorae Flos and Lonicerae Japonicae Flos, the high-frequency materials in this study, reprograms the immune microenvironment and exhibits anti-melanoma effects based on the mechanism that inhibits STAT3 signaling in B16F10 melanoma-bearing mice [129]. Meanwhile, Salvia Miltiorrhizae Radix, another core herb, and Notoginseng Radix et Rhizoma and 6:4 ratio were combined, and synergistic interaction was observed with respect to the protective effect of endothelial cells [130]. These previous studies suggest that rather than predicting the effect of EAHM only on the pharmacological activity of a single herb, considering the interaction between multiple materials together can bring better therapeutic outcomes. From this point of view, as a result of examining the relationship between the core herbs through social network analysis, close connectivity between all materials and an almost uniform level of betweenness centrality were observed. This finding supports the assumption that in the EAHM prescription of this study, the core herbs exerted an effect not only on the effect mechanism of individual herbs but also on the prescription composition principle according to the "Gun-Shin-Jwa-Sa" was considered by the application method. Therefore, tracking the synergy effect derived from the combination of key herbs and searching for the optimal herbal combination that can maximize this synergistic interaction can be a goal in follow-up studies for drug candidate proposals.

Limitations and Perspectives
To use the results and hypotheses derived from this study for clinical decision-making or follow-up research, it is necessary to understand the following limitations. First, as a result of performing a meta-analysis, a significant level of heterogeneity was observed. This suggests that it is difficult to accept that all EAHM prescriptions included in this study are useful for psoriasis. To investigate the cause of heterogeneity in detail, in this review, both outlier sensitivity analysis on individual trials and meta-regression on pre-specified moderators were performed. As a result, in the case of PASI 60, it was found that the type of CM adopted as an active control could be the cause, but in the case of a continuous PASI score showing a higher heterogeneity, a specific cause could not be identified. After excluding other causes, it could be presumed that the high heterogeneity was due to the very diverse composition and dosage of the EAHM prescription in each included trial. A similar problem is often seen in other meta-analyses of EAHM. This is due to EAHM's prescription principle, which requires personalized prescription of herbal materials, and is highly likely to be repeated in future studies of the same design. Additional analysis of herbal material using data mining was performed as a way to overcome the essential limitations due to the characteristics of the intervention itself. If it is not a study that determines only natural products produced by pharmaceutical companies as the scope of analysis, it is thought that a separate analysis of EAHM prescriptions and herbal constituents by various methods in a systematic review related to EAHM in the future will be essential. Second, the commonly prescribed herb-related results derived from this review merely narrowed the scope of hypotheses about core materials through descriptive statistics and unsupervised learning techniques. Therefore, verification of whether the identified core herbs exert a better effect on psoriasis by themselves and whether actual synergy is created from the observed close correlation should be conducted through separate follow-up research. Based on the hypothesis presented in this study, it is thought that useful candidates can be further narrowed by comparing the effects of each EAHM through the network meta-analysis or predicting the mechanism using the network pharmacology technique together with the laboratory research. Third, as the primary outcome in this study, PASI 60 and PASI 70, which were adopted in the most inclusive studies, were selected as a relatively validated endpoint. However, considering that the evaluation instrument used as international standards in recent years is PASI 75 or PASI 90, it is difficult to completely rule out the bias in the results of this study as well. Therefore, in order to more objectively evaluate the efficacy of EAHM compared to placebo control as well as an active control, studies using widely used standard endpoints should be conducted. Fourth, most of the clinical trials included in this review lack pre-registered protocols, do not adopt double-blind methodologies and do not describe detailed randomization procedures. This shows that a number of studies cannot dissipate qualitative concerns, which will also affect the reliability of the results. Although the quantitative growth of EAHM-related evidence over the past decade has been remarkable, more clinical trials are still needed to ensure qualitative progress. Finally, a limitation is that all trials included in this study were conducted in China. In the process of collecting the literature for systematic review, there was no language restriction, and both databases in East Asia, as well as English databases, were searched, but only studies conducted in China met the inclusion criteria. However, as mentioned above, EAHM is widely used as a drug with a common material throughout East Asia, and the academic theory that is the principle of the application is also shared. Therefore, it is considered that the imbalance of trial execution regions is only due to differences in the medical research environment in each country. Therefore, it is expected that this difference can be overcome by continuously conducting studies such as this review on the usefulness of EAHM.

Conclusions
This systematic review supports that oral EAHM monotherapy can be a useful treatment for inflammatory skin lesions in psoriasis. Meta-analysis showed that EAHM had superior effects compared to the control group in PASI 70, PASI 60, continuous PASI score, IL-17, TNF-α, and DLQI of psoriasis patients. In addition, EAHM decreased the incidence rate of adverse events compared to the CM control group. In other words, it is thought that EAHM can positively contribute to skin symptoms, inflammatory status, quality of life, and drug adherence in psoriasis patients.
Further analysis of the EAHM prescription identified 16 high-frequency key materials: Rehmanniae Radix Recens; Salviae Miltiorrhizae Radix; Glycyrrhizae Radix et Rhizoma; Moutan Radicis Cortex; Lithospermi Radix; Smilacis Rhizoma; Radix Paeoniae Rubra; Dictamni Radicis Cortex; Imperatae Rhizoma; Hedyotidis Herba; Isatidis Radix; Lonicerae Flos; Sophorae Flos; Scutellariae Radix; Forsythiae Fructus; Spahalobi Caulis. They are generally thought to show multipath anti-inflammatory activity based on "heat clearing" action and show close connectivity. Therefore, in drug discovery related to this topic in the future, it is expected that the maximization of the anti-inflammatory synergy effect by the combination of EAHM materials belonging to the "heat clearing" category can be treated as a useful research hypothesis.
Despite the above results, concerns about the quality of the included studies and various biases were detected. To reach a firmer conclusion, additional clinical trials that include a multicenter design, a double-blind method, and an outcome with more validity in the design will need to be conducted in the future.