Abstract
Background
Alzheimer's disease (AD) is a progressive neurodegenerative disorder whose global prevalence continues to rise, yet treatment options are still limited. Natural medicines, with their potential for multi-target intervention, have become a key direction in AD drug development. However, a systematic overview of research trends in this field based on bibliometric methods is currently lacking.
Objective
This study aims to summarize research progress on natural medicines for AD treatment using bibliometric analysis and to identify future research hotspots and trends.
Methods
Relevant publications were retrieved from the Web of Science Core Collection. Data visualization and analysis were conducted using VOSviewer, CiteSpace, and R.
Results
A total of 3800 publications were included, involving contributions from 108 countries/regions, 4024 institutions, 18,729 authors, and 706 journals. Publication output showed steady growth, with China and India as the leading contributing countries. Institutions such as the Chinese Academy of Sciences and Kyung Hee University demonstrated high productivity and influence. The research focus has shifted from initial clinical pharmacology and molecular pathology to exploring multi-target mechanisms of natural medicines through network pharmacology and molecular docking. Promising natural agents include Ginkgo biloba, ginseng, curcumin, resveratrol, and Centella asiatica.
Conclusions
Research on natural medicines for AD has progressed steadily over the past two decades, with current emphasis on elucidating multi-target mechanisms using emerging technologies. However, clinical evidence remains limited. Future studies should strengthen multi-omics integration and clinical translation to foster innovative AD prevention and treatment strategies.
Introduction
Alzheimer's disease (AD) is a progressive and fatal neurodegenerative disorder and the most common form of dementia. Currently affecting approximately 44 million people worldwide, AD is projected to affect over three times as many individuals by 2050 due to a growing aging population.1,2 The disease is characterized clinically by progressive decline in cognitive domains such as memory, executive function, and language ability. Key pathological features include amyloid-β (Aβ) deposition, neurofibrillary tangles formed by hyperphosphorylated tau protein, neuroinflammation, and synaptic dysfunction. These processes ultimately lead to widespread neuronal loss and neural network failure.3,4 Despite decades of preclinical research, the translation of novel therapeutic strategies into clinical use remains slow, with high attrition rates in clinical trials underscoring the gap between laboratory findings and effective treatments. Symptomatic treatments for AD, such as cholinesterase inhibitors (e.g., donepezil, rivastigmine, galantamine) and the NMDA receptor antagonist memantine, can temporarily alleviate cognitive and functional symptoms. Although generally well-tolerated in outpatient settings, these drugs do not alter the underlying disease process. 5 In contrast, monoclonal antibodies targeting Aβ plaques represent the first drug class approved to delay clinical progression in early AD. The U.S. Food and Drug Administration (FDA) has granted accelerated approval to lecanemab and full approval to donanemab. 6 While these therapies statistically slow cognitive decline, they carry substantial risks, including amyloid-related imaging abnormalities (ARIA) that may lead to edema or hemorrhage. As a result, treatment requires careful patient screening, administration in a monitored clinical setting, and regular MRI safety evaluations.7,8 Notably, the accelerated approval previously granted to aducanumab has since been withdrawn by the manufacturer. 8 In addition to amyloid-based therapies, several trials are investigating the role of tau-targeting agents, anti-inflammatory compounds, and metabolic modulations as potential disease-modifying treatments. Despite the promising nature of these therapies, challenges such as variability in patient response, treatment-related adverse events, and difficulty in recruiting participants for large-scale trials have hindered the rapid translation of preclinical findings into effective clinical treatments. Consequently, developing novel treatment strategies that can modify the fundamental disease pathology remains a paramount research objective.
In this context, natural products have been explored as a potential avenue for novel AD treatments, although clinical evidence remains limited and translation to human therapies has been slow. With a long history of use in traditional medical systems across Europe, China, South Korea, and India, natural medicines have been widely valued and utilized.9,10 These compounds often exhibit multi-target mechanisms, mild side effects, and sustained efficacy, underscoring their considerable therapeutic potential. 11 It is important to note that while preclinical studies show promise, the current state of clinical trials for natural products in AD remains in the early stages, with few products advancing to large-scale human trials. Notably, the natural product-derived agent sodium oligomannuronate (GV-971) has shown promise in Phase III clinical trials in China, with reported benefits on cognitive function, and is under investigation in global studies, 12 signaling a transition from traditional applications toward modern evidence-based drug development. Furthermore, the World Health Organization's 2014–2023 global strategy to evaluate the safety, efficacy, and quality of herbal medicines has encouraged their integration into national health systems, highlighting the growing recognition of natural products—such as Ginkgo biloba extract (EGb761), Crocus sativus L., Centella asiatica, ginseng, and Rosmarinus officinalis L.—which have demonstrated anti-inflammatory effects and the ability to ameliorate memory deficits in AD clinical and preclinical studies.13–18 Similarly, bioactive natural molecules such as curcumin, 19 quercetin, 20 resveratrol, 21 huperzine A, 22 ginsenoside Rg1, 23 and rosmarinic acid 24 have also shown promising biological activities. Despite these encouraging findings, the research landscape remains fragmented. The scale, developmental context, and collaborative patterns within this field are not yet well understood. This underscores the need for a comprehensive bibliometric analysis to systematically map research trends, identify core contributors, and illuminate future directions in the application of natural products for AD therapy.
Bibliometric analysis offers a distinct advantage over traditional narrative reviews by employing quantitative methods to examine large volumes of scientific literature. Using tools such as CiteSpace and VOSviewer, this approach enables the visualization of research landscapes, revealing key contributions from institutions, countries, and authors, identifying core journals and pivotal publications, and tracing the evolution of research hotspots and emerging trends. 25 Although interest in natural medicines for the treatment of AD continues to grow, a systematic bibliometric assessment of this field remains lacking. To address this gap, this study conducted a quantitative analysis of relevant publications from 2000 to 2023, intending to map the development trajectory, current research fronts, and major trends in natural drug research for AD. The findings are intended to provide valuable insights to guide future drug development and scholarly efforts in this area.
Methods
Data collection and retrieval strategy
Data for this study were retrieved from the Science Citation Index Expanded (SCI-Expanded) database within the Web of Science Core Collection, curated by Clarivate. The search period spanned from January 2000 to June 2025 to systematically identify relevant literature on natural drugs in AD research. The following search strategy was applied using topic terms: TS = ((“Natural medicine” OR “Natural drug” OR “Traditional Chinese medicine” OR “Chinese medicine formula” OR “Herbal medicine” OR “Medicinal plant” OR “Plant extract” OR “Herbal compound” OR “Phytochemical”) AND (“Alzheimer*”)). 26 This search returned 4412 records. To enhance data accuracy and relevance, the literature types were restricted to original research articles and reviews. Book chapters, conference abstracts, retracted publications, and other non-research materials were excluded. The resulting corpus of publications was subjected to further rigorous screening and analysis.
Literature screening criteria
Following the initial filtration by publication type, a dual-independent reviewer system was implemented to ensure objectivity and accuracy in the screening process. Two researchers (Zhao Ran and Liang Chen) independently evaluated the titles and abstracts of all retrieved records to determine whether the studies addressed the application of natural drugs in AD research. Any discrepancies between the reviewers were resolved through discussion with the other two researchers (Wenjun Wang and Tao Zhu) until consensus was achieved.
After this rigorous screening process, a total of 3800 publications met the eligibility criteria and were included in the analysis (Figure 1). The included literature encompasses key research areas regarding natural medicines in AD, including: natural product-based treatment strategies, identification of therapeutic targets, development of in vitro cellular models, and validation through in vivo animal experiments.
Flowchart of the bibliometric analysis process.
Data extraction, analysis, and visualization
The full metadata of all included publications were exported in plain text format from the Web of Science for subsequent bibliometric analysis and visualization. The analysis was conducted using VOSviewer (version 1.6.19), CiteSpace (version 6.3.R1), and Microsoft Office Excel and PowerPoint (version 2019). VOSviewer and CiteSpace were employed to perform a range of bibliometric tasks, including co-occurrence, collaboration, and citation analysis, based on their respective built-in algorithms. In the resulting knowledge maps, nodes represent elements such as keywords, authors, or cited references, with their size corresponding to frequency of occurrence, citation count, or co-citation strength. Links between nodes indicate relationships, and the thickness of each link reflects the strength of association.
Results
Analysis of annual publications
Based on a topic-specific search of the Web of Science Core Collection, this study identified literature on natural drugs and AD published between January 1, 2000, and June 1, 2025. After excluding irrelevant publications, a total of 3800 articles were included, comprising 2755 research articles and 1045 reviews (Figure 2A). Analysis of subject categories revealed that “Pharmacology & Pharmacy” (1202 articles, 31.6%), “Chemistry, Medicinal” (700 articles, 18.4%), and “Biochemistry & Molecular Biology” (668 articles, 17.6%) were the most prevalent fields (Figure 2B). Publication output demonstrated consistent growth from 2000 to 2023, reaching a cumulative total of 3800 articles (Figure 2C). The temporal trend can be divided into three phases: an initial period of slow growth (2000–2009) with fewer than 60 articles per year; a phase of stable expansion (2010–2016), during which annual publications rose gradually to 160; and a recent period of rapid growth (2017–2023), marked by a substantial increase in output that peaked at 418 articles in 2022. This sustained rise in annual and cumulative publications reflects growing academic interest and intensifying research activity in the field of natural drugs for AD. It should be noted that the growth trend shown in this paper is based on the absolute number of publications, and it should be noted that this trend is also affected to some extent by the overall expansion of the scale of global scientific publishing.
Ad natural medicine publication trend. (A) Distribution of literature types; (B) Subject category distribution; (C) Trend of annual publication volume.
Analysis of countries and institutions
The analysis of contributing countries and institutions is based on 3800 publications spanning 108 countries and 4024 institutions, reflecting broad global engagement in this field. As shown in Table 1, China holds a dominant position with 1585 publications, accounting for 41.71% of the total output. India and the USA rank second and third, contributing 484 (12.74%) and 326 (8.58%) articles, respectively. Together, China and India account for more than half of all publications, underscoring the substantial contribution of Asian countries to natural product research in AD. Notably, while the USA ranks third in total publication count, it leads in research influence, as evidenced by an average citation rate of 61.04 per article—significantly higher than that of other countries. The USA also exhibits strong international collaboration, with a total link strength of 369, ranking second globally. Additionally, Italy and Iran demonstrate considerable academic impact, with high average citation rates of 51.03 and 44.57, respectively, indicating that their publications have attracted significant scholarly attention.
Top 10 productive countries related to natural medicine in AD.
At the institutional level (Table 2), seven of the top ten contributing institutions are based in China. The Chinese Academy of Sciences leads with 85 publications and an average of 37.64 citations per article, placing it among the top-performing institutions globally. Beijing University of Chinese Medicine and Kyung Hee University follow with 82 and 70 publications, respectively. Notably, Selçuk University in Türkiye, though producing fewer publications (52), achieved a high average citation rate of 35.75 per article, exceeding that of many high-output institutions, reflecting its strong academic influence. Conversely, some institutions, such as China Medical University, exhibit relatively low total link strength, suggesting more limited international collaboration. To further elucidate patterns of scientific cooperation, we constructed two collaborative networks: a country-level network (Figure 3A), which includes 40 countries with at least 18 publications each, and an institution-level network (Figure 3B), comprising 56 institutions each with a minimum of 20 publications. These networks visually represent the structure and intensity of collaboration and knowledge exchange at both macro and micro levels.
Country and institution collaboration networks. (A) Country-level collaboration network; (B) Institution-level collaboration network. Node size corresponds to the number of publications associated with each country or institution, while connecting lines represent collaborative relationships between them.
Top 10 productive institutions related to natural medicine in AD.
Analysis of journals and co-cited journals
Publications related to natural medicine research in AD are distributed across 706 academic journals. As shown in Table 3, the Journal of Ethnopharmacology leads in number of publications with 197 articles (5.18%), followed by Frontiers in Pharmacology (152, 4.00%) and Molecules (126, 3.32%). Among these high-output journals, Phytomedicine holds the highest impact factor (IF = 8.3), followed by Biomedicine & Pharmacotherapy (IF = 7.5) and Neural Regeneration Research (IF = 6.7). Notably, Phytotherapy Research achieved the highest average citation count per article (65.16), significantly surpassing other journals, which indicates the considerable academic influence of articles published in this journal.
Top 10 productive journals related to natural medicine in AD.
We selected 40 journals that had published at least 21 relevant articles to construct a journal collaboration network (Figure 4A). This network illustrates scholarly connections among journals and reflects increasing interdisciplinary integration in this research area. Among these, Frontiers in Pharmacology has been particularly active in recent years. In terms of highly co-cited journals (Table 4), where “co-cited journals” refer to journals that are frequently cited together within the same reference lists, thereby revealing their thematic relatedness and collective influence within a research field, the Journal of Ethnopharmacology ranks first with a total of 6228 citations, substantially surpassing other journals. It is followed by Molecules (4138 citations) and the Journal of Alzheimer's Disease (3207 citations). This co-citation analysis helps identify core knowledge sources and map the intellectual structure of the field. By filtering for journals with at least 1000 citations, we identified 45 key journals and constructed a journal co-citation network (Figure 4B) to visualize their relationships and roles within the knowledge structure of the field.
Journals in AD natural medicine research. (A) Journal network: the node size represents the frequency of journals, and the connection means the cooperation between journals; (B) Journal co-citation network: the node size represents the frequency of co-citation, and the connection reflects the co-citation correlation between journals.
Top 10 most co-cited journals in natural medicine for AD.
Based on the comparative analysis above, the Journal of Ethnopharmacology ranks first in number of publications, total link strength, and co-citation frequency, underscoring its central role in this field. It serves not only as a key platform for publishing research findings but also as a major source of knowledge and references. Other journals also exhibit distinctive profiles. For instance, despite its relatively lower impact factor (IF = 4.6) and Q2 JCR ranking, Molecules demonstrates a high volume of publications and strong co-citation frequency, indicating its consistent publication of influential and widely referenced studies. Overall, research on natural medicines for AD displays notable interdisciplinary characteristics, integrating pharmacology, biochemistry, neurology, and traditional medicine, reflecting a collaborative and convergent academic landscape.
Analysis of authors and co-cited authors
A total of 18,729 authors have contributed to the field of natural medicine for AD research. As shown in Table 3, the top five authors each published at least 18 articles, with Zengin Gokhan leading (47 articles), followed by Wang Qi (37 articles) and Li Hao (20 articles). A co-authorship network was constructed based on authors with a minimum of 10 publications (Figure 5A). The results reveal limited overall collaboration among authors, with three major clusters formed around Zengin Gokhan, Wang Qi, and Wang Ping. Notably, four of the top five authors belong to Wang Qi's collaborative network, while the remaining authors exhibit relatively independent collaboration patterns. Figure 5B illustrates the relationship between authors’ citation counts and publication years. Sharifi-Rad Javad received the highest total citations (1923), with an average of 192.3 citations per article—despite a relatively recent publication timeline. Zengin Gokhan follows with 1458 total citations and an average of 31.02 per article, while Li Min garnered 950 total citations, averaging 52.78 per article. These metrics indicate that Sharifi-Rad Javad, although producing fewer publications, has achieved notable academic impact and maintains active research collaboration.
Author collaboration co-citation network. (A) Author cooperation network: the size of the node reflects the number of papers published by the author, and the connection reflects the cooperation between the authors; (B) Author citation network: The node size represents the author's citation frequency, and the thickness of the connection reflects the strength of the co-citation correlation. (C) Author co-citation network: node size represents the frequency of co-citation, and the connection demonstrates the cooperation between authors.
Co-cited authors, defined as two or more authors who are cited together by subsequent publications, help uncover underlying thematic associations, identify potential collaborative relationships, and trace the evolution of academic thought within a field. 27 Among the 106,852 co-cited authors identified, five were cited more than 330 times (Supplemental Table 1). Ellman GL received the highest number of co-citations (478), followed by Selkoe DJ (325). Figure 5C presents a network of authors co-cited at least 140 times, illustrating active intellectual exchange and scholarly recognition among influential researchers.
Co-cited reference analysis and burst detection
Over the past two decades, this study identified a total of 185,654 co-cited references related to the use of natural drugs in AD research. Among these, five articles were cited more than 100 times (Supplemental Table 2). The most frequently cited work was a 1961 paper by Ellman et al. in Biochemical Pharmacology, which introduced a method for detecting acetylcholinesterase activity. 28 With 434 citations, this paper significantly surpassed all others, underscoring its foundational contribution to AD research methodology. A co-citation network was constructed using references cited at least 60 times (Figure 6A). As illustrated in Figure 6A, key nodes such as “Ellman GL, 1961, Biochem Pharmacol”, “Hardy J, 2002, Science” (which advanced the amyloid hypothesis and its therapeutic implications), 29 and “Widmann CN, 2015, Lancet Neurol” (which explored the role of neuroinflammation in AD) 30 exhibit strong co-citation associations. This not only reflects the significant impact these publications had at the time of their release but also demonstrates their continued relevance in providing a theoretical foundation and methodological guidance to the field, highlighting their enduring academic vitality.
Co-citation network and burst analysis. (A) Co-citation network. Node size corresponds to the co-citation frequency of a reference; connecting lines reflect the strength of co-citation relationships. (B) Top 25 references with the strongest citation bursts.
Additionally, burst detection analysis of references was performed using CiteSpace to identify publications that experienced a sudden increase in citation frequency over specific time periods. Figure 6B displays the top 25 articles with the strongest citation bursts. Each segment in the figure represents one year, with the red portions indicating the burst period. In terms of burst strength, the articles by Scheltens P (2021) and Ferreira-Vieira TH (2016) showed the highest burst values (17.34 and 16.12, respectively), reflecting their considerable recent influence within the field. Among these, the 2021 review by Scheltens P represents a notable recent contribution. 31
Highly cited articles
Among the 3800 articles retrieved, the top ten by total citation count comprise three research articles and seven reviews (Supplemental Table 3). The most cited research article, with 1232 citations, was published by Lim GP et al. in the Journal of Neuroscience in 2001. 32 This study demonstrated that curcumin significantly enhances anti-inflammatory responses in an AD mouse model, reduces expression of the astrocyte marker GFAP, and decreases insoluble Aβ, soluble Aβ, and plaque burden by 43–50%. The most frequently cited review received 1030 citations and was published by Anwar F et al. in Phytotherapy Research in 2007. 33 This article systematically examines the anti-inflammatory and antioxidant properties of various parts of the medicinal plant Moringa oleifera (leaves, seeds, roots, flowers, and bark). Together, these highly cited works underscore the significant influence of natural products—such as curcumin and moringa—in AD research, particularly highlighting the central role of anti-inflammatory and antioxidant pathways in mitigating disease pathology.
Keyword analysis
Keyword co-occurrence analysis serves as a vital method for identifying core research topics within a specific field. In the resulting network, nodes represent keywords, with their size proportional to frequency of occurrence, while connections between nodes indicate co-occurrence relationships. As illustrated in Figure 7A, the close and complex structure formed among keywords reflects the multi-faceted and interconnected research themes surrounding natural drugs in AD research. Figure 7B displays the temporal distribution of keywords. Emerging terms such as “neuroinflammation,” “anti-inflammatory,” “network pharmacology,” and “molecular docking” have gained prominence since approximately 2021, highlighting a shift in research focus toward inflammatory mechanisms and computational pharmacology approaches in this domain. Alzheimer's disease” emerges as the core keyword with 1294 occurrences and a link strength of 3,392, substantially higher than other terms, underscoring its central position within the research landscape. High-frequency keywords such as “oxidative stress,” “neuroprotection,” “antioxidant,” and “traditional Chinese medicine” reflect prominent research themes, including mechanisms of oxidative damage, neural protection strategies, and the efficacy of traditional herbal treatments.
Knowledge map AD natural drug keywords. (A) Keyword co-occurrence network; (B) Timezone overlay visualization of keyword occurrence; (C) Keywords with the strongest citation bursts.
Keyword burst detection identifies terms that experience a significant surge in frequency over a specific period, thereby revealing emerging hotspots, evolving trends, and shifting research frontiers within a field. 34 As shown in Figure 7C, among the top 20 keywords with the strongest citation bursts, “dementia” had the earliest burst onset, beginning in 2000 and lasting 11 years, with a burst strength of 11.46. The keywords with the shortest burst duration were “induced apoptosis” and “rats,” each lasting 6 years, while “pc12 cells” showed the highest burst intensity at 12.49. The evolution of keyword bursts displays distinct chronological trends. Early research (2000–2010) emphasized clinical features and pharmacological interventions, reflected in keywords such as “dementia,” “double-blind,” and “acetylcholinesterase inhibitors,” indicating a focus on dementia diagnosis and clinical trials of cholinesterase inhibitors. In the subsequent period (2004–2014), investigations deepened toward cellular and molecular mechanisms, with emerging keywords including “pc12 cells,” “lipid peroxidation,” “amyloid beta protein,” and “peptide,” highlighting research interests in oxidative stress, Aβ toxicity, and cellular models. Between 2008 and 2019, attention expanded to cognitive and behavioral domains, as seen in bursts of terms like “learning and memory,” “long term potentiation,” and “psychological symptoms.” More recently (2018–present), the research frontier has shifted toward natural products and their mechanisms, with keywords such as “natural-products,” “docking,” and “resveratrol” indicating growing interest in multi-target therapies, computational drug screening, and bioactive natural compounds.
Discussion
Based on VOSviewer and CiteSpace software, this study conducted a bibliometric analysis of the literature on the treatment of AD with natural drugs in the core collection of Web of Science from 2000 to 2025, and systematically sorted out the characteristics of annual publication trends, country/region distribution, institutional cooperation, journal publication, author co-occurrence, highly cited literature, and keyword hotspots. This analysis is helpful to grasp the international research situation in this field as a whole and provide a reference for the follow-up research direction.
Research overview and publication characteristics
In the past two decades, the research on natural drugs for AD has shown a sustained and significant growth trend. This growth is not an isolated phenomenon but echoes the increasing burden of AD disease in the context of global aging and the increasing attention of natural products in multi-target treatment strategies for complex diseases. The number of annual publications has grown from a slow accumulation in the early years (2000–2009), through a period of steady expansion (2010–2016), to a rapid growth in the recent period (2017–2023), and to a peak in 2022. This trajectory clearly reflects the increasing academic attention and research activities in this field. It is foreseeable that the quantity and quality of relevant research results will continue to rise in the next few years.
Analysis of national and regional collaboration indicates that China holds a leading position in research on natural medicines for AD treatment, representing the most prolific contributor in this field, followed by India, the United States, and South Korea. Notably, although not ranking in the top four by overall publication volume, Iran and Turkey have made significant and longstanding contributions to the field, particularly in the exploration of herbs and spices for AD treatment. For instance, research from Iran has significantly advanced the modern clinical validation of Crocus sativus (saffron). A series of randomized double-blind controlled trials led by Professor Akhondzadeh demonstrated that saffron extract is equivalent to the standard drug donepezil in improving cognitive function in patients with mild-to-moderate AD, with a comparable safety profile, 35 and is significantly more effective than placebo. 36 These high-quality studies provide a valuable model for translating traditional medicinal plants into evidence-based therapies.15,37,38
Furthermore, while the United States ranks third in total publication output, the average citation frequency of its papers is significantly higher, highlighting the substantial impact of its research. The US, along with Italy and other countries, also demonstrates strong international collaboration, which is often a key driver of high-quality and innovative research. At the institutional level, leading Chinese institutions such as the Chinese Academy of Sciences excel not only in productivity but also in average citation frequency, indicating a balance between quantity and impact. Nevertheless, the relatively limited international collaboration networks of some other high-output institutions suggest potential for enhancing cross-border and interdisciplinary cooperation in the future. The analysis of the author collaboration network reveals several research clusters centered around scholars such as Zengin Gokhan, Wang Qi, and Wang Ping. While collaboration within these teams is close, cross-team cooperation remains relatively limited. To foster breakthroughs in natural medicine research for AD, it is crucial to encourage scholars to transcend team and regional boundaries and engage in deeper, more open collaborative innovation.
A total of 3800 relevant articles were published across 706 academic journals. By publication volume, the Journal of Ethnopharmacology ranks first, followed by Frontiers in Pharmacology and Molecules. Notably, the Journal of Ethnopharmacology also occupies the leading position in the journal co-citation network. These consistent findings demonstrate that the journal not only leads in output quantity but also holds a central role in scholarly communication and knowledge linkage, underscoring its significant academic influence and leadership within the field of natural medicine research for AD.
Research trends and hotspots in natural medicines for AD
In-depth keyword analysis reveals a clear evolution in the focus of natural medicine research for AD. Early studies centered on acetylcholinesterase inhibitors, Aβ production and toxicity, and oxidative stress. The field has since expanded to include synaptic function regulation, systematic exploration of traditional medicines and natural products, and the adoption of advanced technologies to accelerate drug development and discovery. In recent years, high-frequency keywords such as “phytochemicals,” “molecular docking,” and “network pharmacology” have consistently emerged, reflecting current research hotspots and signaling a shift toward multi-target drug discovery, computational screening, and mechanistic elucidation of natural compounds.
Additionally, analysis of the terminology indicates that animal models, specifically rodents such as rats, are notably prevalent in the field of anti-AD research. Consequently, a large portion of the current findings are not based on human data, underscoring the translational gap between species.
Commonly used natural medicines for AD
Ginkgo biloba
Ginkgo biloba (GB) was identified 80 times in the keyword frequency analysis, reflecting its considerable attention within AD research. As one of the world's oldest medicinal plants, GB has been used in traditional Chinese medicine for over 5000 years, historically applied in the treatment of respiratory conditions such as asthma. 39 Its bioactive constituents are diverse, primarily including terpenoids (e.g., ginkgolides A, B, and C), flavonoids (e.g., quercetin, kaempferol, and isorhamnetin), as well as polyphenols and organic acids. These compounds have been widely reported in both preclinical and clinical studies to exhibit multiple pharmacological activities, including anti-inflammatory, antioxidant, and anti-apoptotic effects.40,41 In AD preclinical studies, GB extract has demonstrated the ability to mitigate mitochondrial oxidative stress-induced damage to hippocampal neurons. 42 The specific extract GBE50 has also been shown to improve cerebrovascular dysfunction and cognitive decline in AD mouse models. 43 Furthermore, AD preclinical studies suggest that GB may confer neuroprotective benefits through modulation of the PI3 K/AKT/NF-κB signaling pathway. 44
Clinically, a review of 15 trials indicated that GB extract, particularly EGb 761, improved cognitive function, neuropsychiatric symptoms, and activities of daily living in patients with certain dementias across 11 clinical studies, with notable improvements in Mini-Mental State Examination scores and neuropsychiatric inventories. However, four clinical studies reported no significant differences compared to placebo. 45 In summary, GB represents a multi-component, multi-target natural product with potential therapeutic benefits in AD, primarily through regulation of oxidative stress, neuroinflammation, and apoptosis. Nonetheless, current evidence remains largely preclinical, highlighting the need for deeper mechanistic investigation and more robust clinical trials to support its translation into effective treatment strategies.
Ginseng
Ginseng appeared 35 times in the keyword frequency analysis, indicating significant research interest in its role in AD. Derived from the roots and rhizomes of Panax ginseng C. A. Meyer, ginseng has been used for thousands of years in China, Japan, and Korea as a valued traditional medicine. 46 Its complex composition includes ginsenosides, polysaccharides, amino acids, volatile oils, and polyacetylene compounds. 47 Preclinical and clinical studies suggest these components exert neuroprotective effects through multiple mechanisms, including anticholinergic activity, modulation of synaptic plasticity, interference with Aβ and tau pathology, and anti-inflammatory and antioxidant actions, supporting its potential value in AD treatment. 48
However, while preclinical results are promising, a meta-analysis of 15 randomized controlled trials showed that ginseng significantly improved memory, particularly at higher doses, but did not produce statistically significant benefits in overall cognitive function, attention, or executive function. 49 This discrepancy suggests there may be mechanistic or efficacy limitations in ginseng's application for AD. Future research should focus on refined experimental designs, population stratification, and deeper mechanistic studies to clarify its therapeutic boundaries and translational potential.
Curcumin
Curcumin was identified 23 times in the keyword frequency analysis. This natural polyphenolic compound is derived primarily from the rhizomes of turmeric (Curcuma longa), characterized by its orange-yellow color and lipophilic nature. It exhibits broad bioavailability and diverse pharmacological potential. 50 Preclinical and clinical studies have confirmed that curcumin possesses multiple biological activities, including anti-inflammatory, antioxidant, anticancer, antimicrobial, anti-obesity, cardioprotective, and neuroprotective effects.51,52
In AD research, curcumin demonstrates notable multi-mechanism therapeutic potential. It mitigates pathology in AD mice by suppressing inflammatory factors such as TNF-α, IL-6, and IL-1β, modulating oxidative stress markers including MDA and SOD, and activating the AMPK signaling pathway. 53 Additionally, curcumin upregulates the Wnt/β-catenin pathway and brain-derived neurotrophic factor, promoting neurogenesis and improving cognitive function in AD mouse models. 54 To enhance its blood-brain barrier permeability and bioavailability, researchers have recently developed nanovesicle-based delivery systems for curcumin. In preclinical studies, these novel formulations significantly reduce neuroinflammation (e.g., IL-1β, IL-6, TNF-α), decrease Aβ plaque deposition, inhibit tau hyperphosphorylation, and restore hippocampal neuronal integrity. Cognitive behavioral tests further indicate that such interventions improve memory and exploratory behavior without inducing systemic toxicity. 55 These advances reflect a shift in curcumin research from fundamental pharmacological studies toward delivery technology innovation and practical application, highlighting its promising translational potential in AD therapy.
Although preclinical in vivo experiments have shown that curcumin can enhance cognitive function in AD models, human studies have shown inconsistent results, with benefits limited to specific cognitive domains. 56 A recent meta-analysis indicated that the beneficial effect of curcumin on cognition is more potent in older and Asian participants than in younger and Western ones (with limited effects). 57 Larger, well-designed randomized controlled trials are needed to determine the efficacy and safety of curcumin in cognitive aging.
Other natural medicines
Beyond the natural medicines previously discussed, several others have also garnered significant attention in AD research and appeared frequently in the keyword analysis, reflecting considerable scientific interest. Resveratrol, a natural polyphenol found in grapes, berries, and red wine, is noted for the stability and high biological activity of its trans-isomer—the predominant form in nature. It exhibits a range of pharmacological properties, including antioxidant, anti-inflammatory, anticancer, cardioprotective, vasodilatory, phytoestrogenic, and neuroprotective effects. 58 In AD preclinical studies, resveratrol has been shown to confer neuroprotection through multiple pathways: activation of the deacetylase SIRT1, inhibition of Aβ aggregation, modulation of Tau phosphorylation, reduction of oxidative stress and neuroinflammation, and enhancement of mitochondrial function and autophagy, processes critical to maintaining neuronal health. 21 Although resveratrol is highly safe, some reports indicate that it has limited efficacy for AD patients to a certain extent. More data are needed in the future to verify the effectiveness of resveratrol in treating AD.59,60
Centella asiatica, a prominent medicinal plant in traditional systems, has gained increasing attention for its potential to ameliorate AD-related cognitive decline. Regarded in Ayurvedic medicine as a rasayana, a rejuvenating and regenerative herb, it has been historically used to enhance memory. 61 Preclinical studies in AD models indicate that extracts of Centella asiatica improve cognitive performance, reduce Aβ deposition, and exert antioxidant effects.62,63 Schisandrae Chinensis Fructus is a traditional Chinese medicine rich in lignans. In AD preclinical studies, this herb has been shown to reduce oxidative stress and inflammatory responses in the hippocampus and inhibit acetylcholinesterase activity. 64 In addition, studies have shown that Schisandra can significantly improve the learning and memory ability of AD model rats. 65 These studies illustrate the multi-target and multi-pathway characteristics of natural medicines in preclinical AD treatment, underscoring their broad developmental potential. Nevertheless, robust clinical evidence remains scarce, and further high-quality trials are essential to validate their therapeutic applicability.
Exploring new strategies for potential AD drug therapies
Network pharmacology is a systematic research strategy that integrates multi-target interventions and drug repositioning to synergistically modulate disease-related signaling pathways, offering a promising approach for achieving more precise and effective therapeutic outcomes. 66 In recent years, this strategy has been widely applied to studies of traditional Chinese medicine (TCM), particularly for complex diseases such as AD. For instance, network pharmacology analyses have been used to identify potential targets of icariin, 67 Corydalis rhizome, 68 paeoniflorin, 69 Jin-Si-Wei, 70 and Uncaria rhynchophylla 71 in AD treatment, providing novel insights into the multi-component, multi-target, and multi-pathway synergistic mechanisms of TCM.
By systematically mapping complex interactions between drug components and disease targets, network pharmacology, complemented by computational methods such as molecular docking and virtual screening, enables more efficient drug discovery at the molecular level. These techniques include standard ligand-receptor docking, virtual screening of compound libraries, flexible-receptor docking, active site prediction, and hydration-aware docking. 72 They help predict binding conformations and free energy between small molecules and target proteins, thereby reducing experimental screening costs and timelines. For example, Nair et al. 73 conducted a computational molecular docking study based on Manasamitra vatakam, systematically evaluating the binding affinity of various compounds to AD-related target proteins. Their results indicated that chrysin, convallatoxin, rutin, galangin, glycyrrhizin, isoliquiritigenin, quercetin, and naringenin exhibited high docking scores, suggesting these may be key neuroprotective bioactive components in herbal formulations. In another study, Cheung et al. 74 combined network pharmacology with molecular docking to predict core pathways and targets of the formula Guhan Yangshengjing. Experimental validation in Aβ-impaired SH-SY5Y cells showed that key components, liquiritigenin and ginsenoside Rh4, along with their medicated sera, significantly downregulated BACE1 expression. The integration of network pharmacology and molecular docking is shifting TCM research from experience-based to mechanism-driven paradigms, offering powerful tools for elucidating the systemic therapeutic mechanisms of complex natural formulations.
Limitations
This study has several limitations. First, the literature included in the analysis was limited to works published through June 2025. Since the Web of Science Core Collection (WOSCC) database is continuously updated, some papers published in 2025 that were already available online were not captured. Thus, our findings may not fully reflect the research landscape of that year. Second, due to the technical requirements of CiteSpace and VOSviewer, we were unable to merge multiple databases for analysis. As a result, articles exclusively indexed in other databases, such as PubMed, Embase, and Scopus, might have been overlooked. Third, only English-language publications were considered, which may have led to the omission of relevant studies published in other languages. Nevertheless, given the considerable overlap of literature across major databases and the recognized authority of WOSCC, we believe this bibliometric analysis still provides a reliable overview and reflects the general trends in the field of natural medicine for AD. Fourth, while our analysis maps the foundational research that supports clinical development, the translation of these preclinical findings into human therapies remains an open question. The absence of clinical trial data integration limits our ability to conclude the therapeutic efficacy of natural products in AD patients. Future research should address this gap by linking bibliometric trends with clinical trial outcomes.
Conclusion
This study employed bibliometric methods to systematically review the evolution and current landscape of natural drug research for AD treatment. The results indicate a clear thematic expansion within the field, shifting from early focuses on acetylcholinesterase inhibition, Aβ protein deposition, and oxidative stress toward emerging areas such as synaptic function regulation, systematic exploration of traditional medicines and natural products, and the development of multi-target therapeutic strategies. Analysis also revealed that while China maintains a high volume of research output, international collaboration remains limited and should be strengthened. Future efforts should leverage advanced technologies, such as network pharmacology and molecular docking, to further investigate the molecular mechanisms of natural drugs in AD, clarify the efficacy and safety of bioactive compounds, and systematically characterize their targets. Concurrently, increased emphasis should be placed on translating these preclinical findings into clinical applications through robust translational research. It is necessary to conduct larger-scale trials to determine the efficacy and safety of natural medicines in treating AD. This review outlined key developmental trajectories and emerging trends, offering a theoretical foundation and a reference for researchers engaged in natural product-based drug discovery for AD. Future efforts should include rigorous clinical validation.
Supplemental Material
sj-xlsx-1-alz-10.1177_13872877261451887 - Supplemental material for Bibliometric analysis of natural medicine in the treatment of Alzheimer's disease: Trends, hotspots, and emerging research fields
Supplemental material, sj-xlsx-1-alz-10.1177_13872877261451887 for Bibliometric analysis of natural medicine in the treatment of Alzheimer's disease: Trends, hotspots, and emerging research fields by Tao Wen, Tao Zhu, Lin-Ying Zhou, Zhao Ran, Liang Chen and Wen-Jun Wang in Journal of Alzheimer's Disease
Footnotes
Acknowledgements
The authors have no acknowledgments to report.
Ethical considerations
Not applicable
Consent to participate
Not applicable
Consent for publication
Not applicable
Author contribution(s)
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by grants from the National Natural Science Foundation of China (grant number 82204761) and the Deyang Science and Technology Bureau (grant number 2024SZY120).
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
All data generated or analyzed during this study are included in this published article and its supplemental material.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
