Efficacy and potential mechanism of main active ingredients of Astragalus membranaceus in animals with hepatocellular carcinoma: a systematic review and meta-analysis

Introduction

Hepatic carcinoma (HCC), or liver malignancy, can be classified into primary HCC and secondary HCC. The pathogenesis and etiology of HCC are not fully determined, and the onset and progression of HCC are related to viral infections, like hepatitis B virus (HBV) and hepatitis C virus (HCV) infections, dietary contamination (aflatoxin, alcohol, and cirrhosis) (1). HCC has imposed a substantial health burden globally, and studies have shown that China has an HCC incidence of 45.3% and a mortality rate of 47.1% in worldwide cases. It is projected that the worldwide cancer-related cost will amount to approximately 25.2 trillion US dollars from 2020 to 2050, while HCC accounts for about 6.5% (2). Because HCC develops insidiously with no typical symptoms in the early stage and has progressed to the middle and advanced stages when clinical manifestations occur, patients with early HCC can be treated by surgery, and those in the middle and advanced stages can be treated by liver transplantation, immunotherapy, and local therapy (3). However, patients have a low clinical cure rate and poor prognosis. Hence, it is particularly pivotal to look for a safer and more effective drug treatment strategy.

A study has found that traditional Chinese medicines (TCMs) and TCM extracts have anti-tumor ability (4). Astragalus membranaceus (AM) is a leguminous Astragalus plant, sweet in nature, lukewarm, non-toxic, targeting spleen and lung meridians, Shen Nong’s Herbal Classic states Astragalus has a sweet taste and is slightly warm. AM can treat carbuncle (a suppurative infection) and long-term unhealed sore, help discharge pus, and promote wound healing; AM has purulent and analgesic effects, suitable for suppurative infection and pain symptoms; AM can be used to treat leprosy, that is, similar to leprosy and other serious skin diseases; AM has a therapeutic effect on hemorrhoids and rat fistula (a fistula disease); AM is good medicine for invigorating qi, suitable for weak symptoms caused by qi deficiency. AM has an adjuvant therapeutic effect on a variety of diseases in children, especially suitable for children with weak physiques (5). AM possesses diverse medicinal properties that include immune regulation, anti-inflammation, and anti-tumor. AM treats tumors by suppressing the division and apoptosis of tumor cells or inducing tumor cells to differentiate into normal cells. Studies have proven that AM has various chemical activities, like polysaccharides, flavonoids, amino acids, saponins, pigment glycosides, and alcohols, of which Astragalus polysaccharides (APS) and astragalosides (AS) have significant anti-tumor activity in colorectal cancer (6), esophageal cancer (7), and ovarian cancer (8). Increasing evidence supports the substantial therapeutic potential of AM in HCC, which has been studied in vivo and in vitro (9-11). By regulating pSmad3C/L & Nrf2/HO-1 signaling pathways, it inhibits the multiplication, migration, and invasion of transforming growth factor-1 (TGF-1)-stimulated hepatoma cells to exert anti-hepatoma effects in vitro and in vivo (12). Studies have found that the AM’s main active components play an anti-hepatoma role by arresting the cell cycle, promoting apoptosis, suppressing the growth of HCC xenografts, reducing serum tumor markers in HCC models, and inhibiting the PI3K/AKT/mTOR signaling pathway, and then down-regulating levels of PI3K, AKT, mTOR genes, proteins, and their phosphorylated products in this pathway. However, there has been no preclinical study-based meta-analysis to assess the treatment effectiveness of AM’s main active ingredients in HCC. This study is intended to integrate relevant animal studies and apply meta-analysis to investigate the effect and possible mechanism of the AM’s main active components against HCC, then offer some clinical guidance for HCC treatments. We present this article in accordance with the PRISMA reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2426/rc).

Methods

Search strategies

Eight databases [PubMed, Web of Science, Embase, China Biomedical Literature Database (CBM), Cochrane Library, Wanfang Database (Wanfang), China Science and Technology Journal Database (VIP), and China National Knowledge Infrastructure (CNKI)] were screened to identify animal-related studies between inception and July 2024 to include studies that could be meta-analyzed. The search included the following keywords: AM and HCC. At the same time, references in other relevant articles as well as grey literature were manually selected to find qualified studies. The detailed search strategy is listed in the Table S1.

Inclusion and exclusion criteria

The predetermined inclusion criteria included: (I) participants: animals with HCC; (II) intervention: total extract of Astragalus (TEA), AS, and APS; (III) control: given only any dose of AM as the main active ingredient in the treatment group; (IV) outcome: main outcome measures were alanine aminotransferase (ALT), aspartate aminotransferase (AST), tumor weight, and tumor volume; secondary outcomes were ALT, AST, alkaline phosphatase (ALP), interleukin 2 (IL-2), interleukin 6 (IL-6), and interferon-gamma (IFN-γ); (V) drug dose: minimum dosage of medication <200 mg/kg/d, maximum dosage: ≥200 mg/kg/d.

The pre-specified exclusion criteria included: (I) in vitro studies or clinical trials; (II) non-HCC animals; (III) duplicates; (IV) reviews; (V) conferences & paper presentations; (VI) articles with inconsistent study types (e.g., case reports, etc.).

Data extraction

The extracted literatures were entered into the software Endnote 21. Following the aforesaid inclusion and exclusion criteria, two investigators (Huanhuan Li & Y.F.) independently screened the literatures. First, titles and abstracts were scanned for preliminary literature screening. After excluding the unmatched studies, the full text of the literatures was browsed to determine whether they were finally included. Any controversial issue was solved by the two sides through discussion and joined by a third party when necessary. Two investigators extracted the data independently by using a pre-developed spreadsheet. Extracted data included first author, publication year, basic characteristics of laboratory animals (breed, sex, sample size, body weight), modeling approach, intervention (nature, treatment, dose administered, duration), outcome measures, and inter-group differences. For tumor weight and tumor volume, the data of the highest and lowest doses of the drug were included; for other outcome indicators, the data of the highest dose of the drug were included.

Assessment of risk of bias

Two investigators (Huanhuan Li & Y.F.) independently used the risk of bias tool SYRCLE to separately appraise the quality of the enrolled studies, covering ten areas of bias: associated with selection, detection, performance, loss, bias report, and others. The GRADE (13) method was used to formally assess the reliability of the evidence. The results are presented in Table S2. Any deviations were settled in consultation with a third investigator (corresponding author).

Statistical analysis

Statistical analysis was implemented using RevMan 5.4. Relative risk (RR) and its 95% confidence interval (CI) served as effect size for dichotomous variables of efficacy, and mean difference (MD) and its 95% CI for continuous variables. Continuous data were analyzed by standardized MD (SMD) and its 95% CI, the associated results were pooled using a random-effects model. Significant differences were considered between the intervention and control groups if P<0.05. Heterogeneity was tested across studies by I-squared (I²) statistics. To examine the robustness of results, when heterogeneity was higher (I²>50% and P<0.1), one study was sequentially eliminated using a one-by-one exclusion method, and the meta-analysis was performed again for the rest studies to observe whether the pooled results changed considerably. STATA software was leveraged to draw funnel plots. For outcomes reported in ≥10 included studies, possible publication biases were identified by funnel plots and Egger’s test.

Results

Articles included

In total, 2,654 possible related articles were retrieved from eight databases, comprising 80 from PubMed, 162 Web of Science, 167 Embase, 12 Cochrane, 366 CBM, 704 CNKI, 751 Wanfang, and 412 VIP. After removing 1,211 duplicates, 1,443 articles were left. Next, the remaining 1,443 articles were read wholely, and 1,390 were eliminated based on the aforesaid exclusion criteria. Eventually, 53 qualified articles were included. All eligible studies were released from 1995 to 2024, indicating that the protection of the AM’s main active components against HCC has been widely investigated. The detailed selection process is in Figure 1.

Figure 1 Flowchart of the selection for studies included.

Characteristics of included studies

Fifty-three studies were included [prior to 2010 (16 studies): (14-29); 2010–2020 (25 studies): (12,30-53); 2021–2026 (12 studies): (11,54-64)]; 1,487 animals with HCC were selected, including mice or rats, with 735 in the treatment group and 752 in the model group; 4 studies chose Sprague-Dawley rats (14,15,26,50) (105/1,487, 7.1%); 24 studies chose Kunming mice (16,17,19-25,28-35,38,40-42,47,62,63); 21 studies chose BALB/c mice (12,18,37,39,43-46,48,49,51-54,56-60,62,64) (476/1,487, 32%); 1 study chose BACB/C mice (27) (5/1,487, 0.3%); 5 studies chose C57BL/6 mice (11,23,24,29,56) (228/1,487, 15.3%); 1 study did not specify the mouse type (61) (12/1,487, 0.8%); 1 study used both Kunming and C57BL/6 mice (29); 2 studies used Kunming, BALB/c, and C57BL/6 mice (23,24); 21 studies (12,16,22-24,27,29-32,34,35,38,39,41-43,46,47,54,58) chose both male and female animals; 23 studies (11,15,18,19,21,26,33,36,37,40,44,45,49,50,55-61,63,64) chose male animals; 9 studies (6,14,17,20,25,51-53,62) chose female animals; 10 studies (12,14,29,36,39,49,58,60,61,64) chose mice of 4–6 week old; 1 study (54) didn’t report mouse body weight and age of mice, the remaining studies described the animal weights. In the construction of HCC models, 3 studies (15,26,50) applied dimethylnitrosamine (DMN) models, 9 studies (12,29,37,39,49,58,60,61,64) applied HepG2 hepatoma cell tumor sources, 24 studies (16,17,19-21,25,29-35,38,40-42,44,46-48,55,56,63) applied H22 hepatoma sources; 3 studies (22,24,28) applied HepA hepatoma cell sources; 2 studies (18,54) applied SMMC-7721 hepatoma cell tumor sources; 5 studies (45,57,59-61) applied HuH7 hepatoma cell tumor sources; 1 study (58) applied Hep 3B hepatoma cell tumor sources; 1 study (49) applied BNL-75 hepatoma cell tumor sources; 2 studies (11,64) applied Huh 7 cells infected with LV-NC or LV-KAT 2A or DEN/CCl4/C2H5OHcombination to establish an animal model; 1 study (49) transplanted liver tumor fragments into the murine margin of the liver under the right rib to establish an animal model; 1 study (14) applied diethylnitrosamine (DEN) combined with hepatectomy and 2-acetylaminofluorene (2-AAF) to establish an animal model. The shortest administration period was 2 weeks, and the longest was 16 weeks; 0.3 mg/kg/d was administrated as the lowest dose of the AM’s main active ingredients, and 12,000 mg/kg/d as the highest. Regarding outcome measures, 18 studies (12,18,27,35,37,38,43,44,50,51,54,56-59,61,63,64) reported tumor volume, tumor weights following low-dose drug treatment were reported in 22 studies (16-18,22-25,28,29,32,42-45,49,54,55,57-60,64). By contrast, 11 studies documented tumor weight data for the high-dose drug treatment cohort (18,27,34,35,37,38,40,47,54,56,63) reported tumor weight; 6 studies (11,26,28,30,42,50) reported ALT levels; 5 studies (11,28,30,42,50) reported AST levels; 3 studies (15,26,28) reported ALP levels; 11 studies (12,23,25,30,36,38,41,47,48,50,56) reported tumor necrosis factor-alpha (TNF-α) levels; 8 studies (25,30,31,36,38,41,48,56) reported IL-2 levels; 5 studies (24,25,41,47,50) reported IL-6 levels; 4 studies (23,36,55,56) reported IFN-γ levels; 3 studies (19,41,58) reported Bax; 4 studies (19,41,53,58) reported Bcl-2; 3 studies (46,49,53) reported MMP-2; 12 studies (17,21,25,30-33,38,42,48,55,63) reported thymus indices (TI); 11 studies (17,21,25,30-33,38,42,55,63) reported spleen indices. Some studies recorded mouse body weight, and other related proteins, such as STAT3, caspase-3, etc, while other studies have recorded oxidative stress indicators, like malondialdehyde (MDA) and superoxide dismutase (SOD). Characteristics of the included studies are listed in Table 1.

Quality of included studies

Good randomization methods were used in all 53 studies, and the assessment of bias indicated that the included studies had an unclear risk of bias, with the detailed risk of bias shown in Figure 2. This study used the Cochrane RoB 2.0 and SYRCLE tools to assess the risk of bias for 53 RCTs. In the randomization process domain, 44 studies were judged as ‘low risk’ because the description of the generation of random sequences and the concealment of allocation was clear; 9 studies were judged as ‘uncertain’ because the randomization method was unclear and the concealment of allocation was not implemented. In the deviation from the intervention domain, 53 studies had a low risk due to good compliance. For the data missing domain, 50 studies were of low risk (with a missing rate of less than 5%), and 3 studies had missing information. In terms of outcome measurement domain, 53 studies had a low risk. In the selective reporting domain, 53 studies had a low risk due to complete reporting. Overall, the studies had a high quality in terms of randomization and measurement. However, the issues of data missing and selective reporting need to be addressed, which may affect the robustness of the results. Subsequently, sensitivity analysis was conducted to verify the impact of bias, and future studies should improve the transparency of the methodological report.

Figure 2 Quality evaluation of the included studies. D, day.

Effectiveness

Primary outcomes

Ten studies (12,18,43,44,50,54,57-59,64) reported the low-dose AM for tumor volume, and uncovered that the AM’s main active ingredients could greatly decreased the liver tumor volume relative to the control group (n=140; SMD, −5.03; 95% CI: −6.95 to −3.10; heterogeneity: P<0.001, I²=87%; Figure 3A, GRADE: low certainty); 9 (18,35,37,40,51,54,56,61,63) reported the high-dose AM for tumor volume, and disclosed that the AM’s main active ingredients could greatly decreased the liver tumor volume relative to the control group (n=149; SMD, −5.80; 95% CI: −8.26 to −3.34; heterogeneity: P<0.001, I²=93%; Figure 3B, GRADE: low certainty). Twenty-two studies (16-18,22-25,28,29,32,42-45,49,54,55,57-60,64) recorded the tumor weight, finding that the main active ingredients of low-dose AM could significantly reduce liver tumor weight compared to the control group (n=502; SMD, −2.14; 95% CI: −2.77 to −1.52; heterogeneity: P<0.001, I²=84%; Figure 3C, GRADE: low certainty). Eleven studies (18,27,34,35,37,38,40,47,54,56,63) reported the tumor weight, finding that the main active ingredients of high-dose AM could greatly decreased the liver tumor weight relative to the control group (n=191; SMD, −2.71; 95% CI: −3.53 to −1.89; heterogeneity: p=0.002, I²=71%; Figure 3D, GRADE: low certainty).

Figure 3 Forest plot of effects of main active ingredients of Astragalus membranaceus on HCC animals. (A) Effect of the low-dose main active ingredients of Astragalus membranaceus on tumor volume; (B) effect of the high-dose main active ingredients of Astragalus membranaceus on tumor volume; (C) effect of low-dose main active ingredients of Astragalus membranaceus on tumor weight; (D) effect of high-dose main active ingredients of Astragalus membranaceus on tumor weight. CI, confidence interval; HCC, hepatocellular carcinoma; IV, inverse variance; SD, standard deviation.

Liver function

Six studies (11,26,28,30,42,50) reported ALT levels, indicating that the AM’s main active ingredients significantly lowered ALT levels in HCC animals relative to the control group (n=156; SMD, −3.95; 95% CI: −6.13 to −1.78; heterogeneity: P=0.0004, I²=94%; Figure 4A, GRADE: low certainty). Five studies (11,28,30,42,50) reported AST levels, indicating that compared to the control group, the AM’s main active ingredients could greatly decreased AST levels (n=126; SMD, −5.91; 95% CI: −9.00 to −2.82; heterogeneity: P=0.0002, I²=95%; Figure 4B, GRADE: low certainty). Three studies (20,33,57) reported ALP levels, indicating that the AM’s main active ingredients tended to lower ALP levels in contrast to the control group, yet no notable difference was identified (n=80; SMD, −0.69; 95% CI: −1.80 to −0.42; heterogeneity: P=0.22, I²=81%; Figure 4C, GRADE: moderate certainty).

Figure 4 Forest plot of effects of main active ingredients of Astragalus membranaceus on HCC animals. (A) Effect of main active ingredients of Astragalus membranaceus on ALT; (B) effect of main active ingredients of Astragalus membranaceus on AST; (C) effect of main active ingredients of Astragalus membranaceus on ALP. ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CI, confidence interval; HCC, hepatocellular carcinoma; IV, inverse variance; SD, standard deviation.

Inflammatory factors

Eight studies (25,30,31,36,38,41,48,56) reported IL-2 levels, finding that the main active ingredients of AM could significantly increase IL-2 (n=136; SMD, 2.88; 95% CI: 1.38 to 4.37; heterogeneity: p=0.0002, I²=88%; Figure 5A, GRADE: low certainty). Five studies (24,25,41,47,50) reported IL-6 levels, noting a downward trend in the AM’s main active ingredients relative to the control group (n=95; SMD, 0.93; 95% CI: −2.05 to 3.91; heterogeneity: p=0.54, I²=95%; Figure 5B, GRADE: low certainty). Four studies (23,36,55,56) reported IFN-γ levels, finding that the AM’s main active ingredients greatly lowered IFN-γ levels relative to the control group (n=99; SMD, 4.67; 95% CI: 1.63 to 7.70; heterogeneity: p=0.003, I²=86%; Figure 5C, GRADE: low certainty). Eleven studies (12,23,25,30,36,38,41,47,48,50,56) reported TNF-α levels, finding that the AM’s main active ingredients greatly lowered TNF-α levels relative to the control group (n=236; SMD, 2.00; 95% CI: 0.66 to 3.33; heterogeneity: p=0.003, I²=92%; Figure 5D, GRADE: low certainty).

Figure 5 Forest plot of effects of main active ingredients of Astragalus membranaceus on HCC animals. (A) Effect of main active ingredients of Astragalus membranaceus on IL-2; (B) effect of main active ingredients of Astragalus membranaceus on IL-6; (C) effect of main active ingredients of Astragalus membranaceus on IFN-γ; (D) effect of main active ingredients of Astragalus membranaceus on TNF-α. CI, confidence interval; HCC, hepatocellular carcinoma; IFN, interferon; IL, interleukin; IV, inverse variance; SD, standard deviation; TNF, tumor necrosis factor.

Protein expression

Three studies (19,41,58) reported Bax levels, suggesting that the AM’s main active ingredients could remarkably upregulate the expression of tumor Bax protein in comparison with the control group, (n=42; SMD, 2.89; 95% CI: 0.70 to 5.07; heterogeneity: P=0.010, I²=80%; Figure 6A, GRADE: moderate certainty); 4 studies (19,41,53,58) reported Bcl-2 levels, finding that the AM’s main active ingredients could greatly downregulate tumor Bcl-2 protein expression in comparison with the control group (n=41; SMD, −3.10; 95% CI: −5.89 to −0.31; heterogeneity: P=0.03, I²=84%; Figure 6B, GRADE: moderate certainty); 3 studies (49,51,52) reported MMP-2 levels, finding that there was a trend of decreased MMP-2 protein expression in the AM’s main active ingredients in comparison with the control group (n=77; SMD, −3.42; 95% CI: −6.83 to −0.02; heterogeneity: P=0.05, I²=93%; Figure 6C, GRADE: moderate certainty).

Figure 6 Forest plot of effects of main active ingredients of Astragalus membranaceus on HCC animals. (A) Effect of main active ingredients of Astragalus membranaceus on Bax; (B) effect of the main active ingredients of Astragalus membranaceus on Bcl-2; (C) effect of main active ingredients of Astragalus membranaceus on MMP-2. CI, confidence interval; HCC, hepatocellular carcinoma; IV, inverse variance; SD, standard deviation.

Thymus index

Twelve studies (17,21,25,30-33,38,42,48,55,63) reported TI, finding that the AM’s main active ingredients increased the thymus index relative to the control group (n=292; SMD, 1.42; 95% CI: 0.66 to 2.18; heterogeneity: P=0.0003, I²=86%; Figure 7, GRADE: moderate certainty).

Figure 7 Effect of main active ingredients of Astragalus membranaceus on the thymus index. CI, confidence interval; IV, inverse variance; SD, standard deviation.

Spleen index

Eleven studies (17,21,25,30-33,38,42,55,63) reported spleen indices, finding that the AM’s main active ingredients increased the spleen index relative to the control group (n=272; SMD, 0.81; 95% CI: −0.11 to 1.74; heterogeneity: P=0.09, I²=90%; Figure 8, GRADE: low certainty).

Figure 8 Effect of main active ingredients of Astragalus membranaceus on the spleen index. CI, confidence interval; IV, inverse variance; SD, standard deviation.

Subgroup analysis

Due to the high heterogeneity among the studies, we conducted subgroup analyses on the effects of tumor volume, tumor weight, ALT, AST, spleen index, and thymus index based on the publication year, the type of the main active component of AM, the administration route of the main active component of AM, and the treatment duration (65). The results indicated that the type of the main active component of Coptis chinensis might be the source of the heterogeneity in tumor weight and AST. Other subgroups did not reveal the source of the heterogeneity in transaminase. The results are presented in Table S3.

Sensitivity analysis

Following sequentially precluding each study, we re-analyzed the data, yielding stable results, and no abnormalities were found in the sensitivity analysis.

Publication bias

For the outcomes of the 53 articles included in the study, funnel plots and Egger’s test were applied to identify publication bias (Figure 9). The funnel plots were symmetrical, and there was no significant publication bias based on Egger’s test (Table 2).

Figure 9 Funnel plot of the main active ingredients of Astragalus membranaceus on hepatic carcinoma animal model. (A) Effect of main active ingredients of low-dose Astragalus membranaceus on tumor weight; (B) effect of main active ingredients of high-dose Astragalus membranaceus on tumor weight; (C) effect of main active ingredients of low-dose Astragalus membranaceus on tumor volume; (D) effect of main active ingredients of high-dose Astragalus membranaceus on tumor volume. se, standard error; SMD, standardized mean difference.

Discussion

Summary of the evidence

It’s proved that AM has certain anti-tumor effects, and its main active ingredients have shown anti-tumor effects in HCC animals. The systematic review and meta-analysis offer important evidence support for clinical research and new drug development for HCC. However, there is no pre-clinical study on the efficacy of the AM’s main active ingredients in HCC. This study included 53 studies involving 1,110 animals. The study analyzed all the outcome indicators of the articles, showing that the AM’s main active ingredients reduced the tumor weight and tumor volume, and enhanced liver function, like ALT, AST, and ALP; lowered inflammatory factors, like TNF-α, IL-2, IL-6, and IFN-γ; protein expression, such as Bax, Bcl-2, and MMP-2; and TI, and spleen indices. In the animal model of liver cancer, after administering the main active components of astragalus, changes in the thymus index and cytokines were identified as potential biomarkers for liver cell carcinoma. However, these changes do not directly reflect the therapeutic progress of liver cell carcinoma caused by the main active components of astragalus. This study suggested that the AM’s main active ingredients obviously ameliorated outcome indicators in HCC animals, and its mechanism may be related to suppressing the development and proliferation of cancer cells and improving immune function and inflammatory response in mice.

Possible mechanisms of action

Primary liver cancer, also known as HCC, is malignant tumors that originate from hepatocytes or intrahepatic bile duct epithelial cells, with complex etiology and pathogenesis. Research has found that the main etiology of HCC is viral hepatitis. Approximately 90% of HCC patients in China have suffered from HBV infection. HBV infection induces chronic hepatitis, which leads to liver cirrhosis and further induces the occurrence of HCC. HBV infection and cell signaling pathways are important mechanisms for the transmission of information inside and outside cells, which can interfere with cell signaling pathways, affecting the proliferation, metastasis, and apoptosis of hepatocytes, leading to the occurrence of HCC. Research has found that blocking the STAT3 signaling pathway with shRNA can promote apoptosis in HBV-positive HCC cells and induce cell cycle arrest, thus inhibiting the HCC cell growth (66). Overexpression of miR-520e inhibits the p38MAPK & ERK1/2 signaling pathways by suppressing EphA2, then decreasing HBV replication, and inhibiting tumor growth (67). Additionally, HBV expresses various active proteins. Research has shown that HBx can affect the regulation of non-coding RNA (ncRNA), RNA, and long ncRNA, and participate in epigenetic modifications and DNA repair. Various signaling pathways (like p53, Wnt, and nuclear factor-κB pathways) interact, confirming that HBx promotes the occurrence and progression of HCC (68). The conclusions related to our pathways are all descriptive in nature and do not stem from the results of a meta-analysis. Aflatoxins, liver fibrosis, chemical toxins, fatty liver, and other factors have a close relationship with HCC. Research has found that oxidative stress can promote the infiltration of inflammatory cells and increase protease secretion, which can mediate the oxidation and reduction modification of the β-subunit of protein kinase A, causing the degraded E-cadherin (ECAD) and promoting tumor metastasis of HCC (69). Tumor cells have reactive oxygen species (ROS) levels, which not only participate in apoptosis, necrosis, and intercellular signaling transduction and gene expression but also cause DNA damage, the formation of lipid substances, and the occurrence of cancer. Research has found that the lack of glucose in hepatocytes leads to tricarboxylic acid cycle (TCA) shunting, energy crisis, and excessive oxidative damage, while the forced expression of PCK1 inhibits the growth of liver tumors induced by the oncoprotein yes-associated protein (YAP-5SA) and promotes the TCA ketogenic, oxidative stress, and HCC apoptosis, inhibiting the growth of HCC (70). The increase in reactive oxygen levels, the induction of mitochondrial dysfunction, and increased ATP consumption lead to further inflammatory signals, promoting inflammation and immune-suppressive microenvironments, activating cancer cell apoptosis and necrosis, and accelerating the occurrence of hepatitis, liver fibrosis, liver cirrhosis, and HCC (71). Liver fibrosis is a pathological process of abnormal proliferation of liver connective tissue. Liver fibrosis leads to a decrease in the regenerative and repair abilities of the liver, causing a decrease in liver function. Liver fibrosis is attributable to the imbalance between the formation and degradation of the extracellular matrix (ECM), providing a developmental environment for dysplasia and the ultimate induction of HCC. Research has found that reducing the availability of TGF-β1 through the consumption of galectin 3-binding protein (LGALS3BP) can alleviate liver fibrosis and inhibit the occurrence of HCC (72). AST-IV can target HMGB1 to the liver cancer cells, thereby triggering ferroptosis in them (73,74). By contrast, the absence of dectin-1 aggravates liver fibroinflammation and promotes the occurrence of HCC. Dectin-1 regulates liver fibrosis and HCC by inhibiting the TLR4 signaling pathway, while the reduction of dectin-1-dependent macrophage colony-stimulating factor (M-CSF) expression decreases the expression of TLR4 and CD14, thus limiting the progression of liver fibrosis and the occurrence of HCC (75).

Future prospects

By considering the particularity of animal experiments, the protection of AM’s main active ingredients against HCC and their possible mechanisms need further verification. Gu et al. (76) revealed that microcystin-LR (MC-LR) could suppress the phosphatase 2A (PP2A) activity, activate the Akt and MAPK signaling pathways, promote liver fibrosis, and facilitate the occurrence of HCC. The latest research has found that imbalances and disturbances between the gut barrier and the gut microbiota can affect the gut-liver axis, and the mechanism of action may be that different bacteria may have different ways of stimulating immune cells, and the intestinal microbiota promotes the occurrence of HCC through impaired function of the gut-liver axis (77). The AM’s main active ingredients have a significant anti-HCC effect, with remarkable therapeutic effects, and are expected to bring new treatment options for HCC patients. Our current clinical results do not directly prove that the main active ingredient of astragalus is reasonable for the treatment of human hepatocellular carcinoma. More rigorous clinical randomized controlled trials are needed to demonstrate the effect of the main active components of astragalus on hepatocellular carcinoma in human models.

Limitations

This study has certain limitations. Firstly, the article quality was poor, with some articles not describing randomization methods and not providing a detailed description of the sample’s experimental methods. Secondly, because of the small sample size in some drug studies, this may cause the resulting bias. The number of drugs directly compared in some articles was also small, which may limit the reliability and accuracy of the results. Thirdly, experimental results may be affected by animal stress responses, leading to unreliable results. Fourthly, the studies included in this research are predominantly from regions in China, which may affect the generalizability of the research findings in other populations. There is a large heterogeneity in the results between humans and animals in terms of anatomy, tissue, and immunity. The source of heterogeneity has not been found, which may have a certain impact on the credibility of the results.

Conclusions

The meta-analysis certifies the effectiveness of AM’s main active ingredients in HCC based on preclinical studies. Its protective mechanisms on HCC animals may include alleviating tumor weight and volume, regulating inflammatory responses, and protecting liver functions. Given the obvious heterogeneity and poor quality of the included studies, positive results should be explained prudently. Studies on AM’s main active ingredients in HCC are mostly animal studies, and further clinical trials are required to elucidate the efficacy.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2426/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2426/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2426/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Chidambaranathan-Reghupaty S, Fisher PB, Sarkar D. Hepatocellular carcinoma (HCC): Epidemiology, etiology and molecular classification. Adv Cancer Res 2021;149:1-61. [Crossref] [PubMed]
2. Chen S, Cao Z, Prettner K, et al. Estimates and Projections of the Global Economic Cost of 29 Cancers in 204 Countries and Territories From 2020 to 2050. JAMA Oncol 2023;9:465-72. [Crossref] [PubMed]
3. Liu CY, Chen KF, Chen PJ. Treatment of Liver Cancer. Cold Spring Harb Perspect Med 2015;5:a021535. [Crossref] [PubMed]
4. Sheng F, Yang S, Li M, et al. Research Progress on the Anti-Cancer Effects of Astragalus membranaceus Saponins and Their Mechanisms of Action. Molecules 2024;29:3388. [Crossref] [PubMed]
5. Man L, Xiaobo L. A Brief Discussion on Zhongjing's Use of Astragalus Based on Shennong's Herbal Classic and Synopsis of Golden Chamber. Chinese Journal of Clinical Medicine 2025;17:74-7.
6. Bian Y, Wang G, Zhou J, et al. Astragalus membranaceus (Huangqi) and Rhizoma curcumae (Ezhu) decoction suppresses colorectal cancer via downregulation of Wnt5/β-Catenin signal. Chin Med 2022;17:11. [Crossref] [PubMed]
7. Chen L, Xing D, Guo LR, et al. Formononetin, an Active Component of Astragalus Membranaceus, Inhibits the Pathogenesis and Progression of Esophageal Cancer Through the COX-2/Cyclin D1 Axis. Clin Lab 2023;
8. Zhang J, Liu L, Wang J, et al. Formononetin, an isoflavone from Astragalus membranaceus inhibits proliferation and metastasis of ovarian cancer cells. J Ethnopharmacol 2018;221:91-9. [Crossref] [PubMed]
9. Fang Gong Y, Hou S, Xu JC, et al. Amelioratory effects of astragaloside IV on hepatocarcinogenesis via Nrf2-mediated pSmad3C/3L transformation. Phytomedicine 2023;117:154903. [Crossref] [PubMed]
10. Tian QE, De Li H, Yan M, et al. Effects of Astragalus polysaccharides on P-glycoprotein efflux pump function and protein expression in H22 hepatoma cells in vitro. BMC Complement Altern Med 2012;12:94. [Crossref] [PubMed]
11. Li L. The investigation of the anti-hepatocellular carcinoma effect of Astragalus membranaceus saponin and its regulation of pSmad3C/L and Nrf2/HO-1 signaling pathways. 2022. Available online: https://link.cnki.net/doi/10.26921/d.cnki.ganyu.2022.001098
12. Guo Z. The role of the PI3K/AKT/mTOR signaling pathway in the anti-liver cancer effect of sodium arsenite in combination with astragaloside [Doctoral dissertation]. Inner Mongolia Agricultural University; 2021.
13. Zhang T, Wang H, Xing D, et al. The application and challenge of GRADE in systematic reviews of animal studies. Chinese Journal of Evidence-Based Medicine 2019;19:248-252.
14. Cui R, He J, Wang B, et al. Suppressive effect of Astragalus membranaceus Bunge on chemical hepatocarcinogenesis in rats. Cancer Chemother Pharmacol 2003;51:75-80. [Crossref] [PubMed]
15. Dang S, Zhang Z, Yuan L, et al. Effects of emodin and astragalus polysaccharides on the expressions of GST-P and TGF-β_1 in rats' hepatocellular carcinoma constitution. Journal of Xi’an Jiaotong University (Medical Sciences) 2006;250-3.
16. Huang HS, Huang WT, Wei PY, et al. Experimental study on Astragalus polysaccharide with cisplatin in treating H22 liver cancer. Journal of Guangxi Medical University 2009;26:19-1.
17. Liu MH, Ren MP, Chen JP, et al. Study on antitumor activity of Astragalus saponins. Pharmacology and Clinics of Chinese Materia Medica 2009;68-70.
18. Chunyu L, Zhenlun G, Chongmin D, et al. Experimental study on the antitumor effects of flavonoids of Astragalws complanatus on nude mice with human liver cancer xenografts. Chinese Pharmacological Bulletin 2007;781-4.
19. Shen TH, Zhang XH, Zheng M, et al. Experimental study on the effects of oral administration of Astragalus membranaceus extract on H22 tumor-bearing mice. Chinese Journal of Clinical Oncology 2006;821-4.
20. Wang HQ, Cui NJ, Hu L, et al. Effect of Heat-clearing and Detoxifying Herbs and Tonifying Herbs on Mice Hepatoma Cells Line H22 and Their Immunologic Mechanism. Journal of Guangzhou University of Traditional Chinese Medicine 2006:156-9.
21. Wei C, Tang Q, Liang Z, et al. The effects of flavonoids extracts from semen Astragali complanali on the growth of liver cancer and immune function. Tumor 2009;29:1112-5.
22. Dujuan X, Qiang W, Yan Y, et al. The antitumor effect and mechanism of Astragalus membranaceus total saponins. Chinese Pharmacological Bulletin 2003;823-6.
23. Xu DJ, Chen MZ. The tumor-inhibiting effect and mechanism of Astragalus polysaccharides. Chinese Journal of Hospital Pharmacy 2005;25:923-5.
24. Xu D. Research on the anti-tumor effects and mechanisms of astragalus total extract. 2003. Available online: https://kns.cnki.net/kcms2/article/abstract?v=X0pQrW14MIueiZ9mNcAvNl_SD-jrWovk7peryXH6unUD1aPXzfIcKel8EjloZZGSKLti-ZrZesdZNEP6fRbyKmQHdGTL-BwpUP-AKSYitdO5Iiu3GQOiDsQbdZQXAqiZDj2dvLOQf-IEWjOKojr2qvYn4qwm4f2ly3X194pUuxGY0fWQbQXd_pIyb5vAHcNjCbzeOE4Alq0=&uniplatform=NZKPT&language=CHS
25. Xiao SH, Ren MP, Liu MH, et al. The effect of Astragalus polysaccharide on levels of IL-2, IL-6, IL-12, and TNF-α in tumor-bearing mice. Sichuan Journal of Physiological Sciences 2009;31:2.
26. Yuan LC, Cheng YA, Dang SS, et al. Preventive effects of Emodin and Astragalus polysaccharides on rats with experimental liver cancer. China Pharmacy 2005:574-6.
27. Meiju Z, Zhenqiu G, Faqing T. Observation on the inhibitory effect of Jianpi Shugan Huoxue therapy on the recurrence of human liver cancer cells in nude mice. Guiding Journal of Traditional Chinese Medicine and Pharmacy 2000;31-2.
28. Zhang XL, Li WH, Li LY, et al. Experimental study on the effect of total astragalus extract on anti-tumor. China Modern Medicine 2010;4:29-3.
29. Zhou SY, Wang LX. Experimental study on the anti-tumor effects of Astragalus polysaccharides (APS). Chinese Journal Of Pharmaceutical Biotechnology 1995;02:22-5.
30. Fang AY, Shi XF, Lu YC, et al. Mechanism of astragalus decoction on apoptosis and liver injury in H22 tumor-bearing mice. West China Journal of Pharmaceutical Sciences 2020;35:392-6.
31. Fang AY, Liu XS, Lu YC, et al. Inhibitive effect of total astragalus saponin on tumor and its mechanism in H22-bearing mice. Chinese Journal of Public Health 2017;33:596-8.
32. Cai Y, Wang Q. Research on inhibited activity of ASTs on liver cancer. China Journal of Modern Medicine 2011;21:4258-60.
33. Du W, Wei Y, Dong X, et al. Study on Anti-tumor Metastasis and Anti-tumor Growth Effects of Water Extract from Astragalus mahoschanicus. Chinese Wild Plant Resources 2012;31:24-7+52.
34. Fan SL, Wang ZJ, Ren YL. Cellular immune mechanisms of Astragalus injection on tumor inhibition in H22 tumor-bearing mice. China Medicine 2013;11:76-7.
35. Huang Y, Li L, Xie H, et al. An experimental study on anti -tumor effects of Astragalus injection on H22 cell-bearing mice. Medical Journal of the Chinese People's Armed Police Force 2011;22:562-4.
36. Lin L. Research on the Anti-Liver Cancer Effects and Immune Mechanisms of Astragaloside and β-Elemene. 2011. Available online: https://kns.cnki.net/kcms2/article/abstract?v=X0pQrW14MIsOzaL7d7JaYpg4GpNl9zEmadblzG_BQ5MGxmvk8D1GYykXeIRGsHOsLn3Kpt1CfnT-VgRXfgsWtWlAFryMaP383Dm6wN7dCmBmzjQrumxbfoIpLrWbRbYLCo_XR4szQQAcmgFdPQS2MMed6kTiU2-z4efElkV6MUxW14oB4LYJiKgNmqNvorPjW-xm2A2JWBQ=&uniplatform=NZKPT&language=CHS
37. Luo S, Yin G, Tang D. The effect of astragaloside combined with curcumin on the angiogenesis of orthotopic liver cancer xenograft in nude mice. Chinese Journal of Cancer Prevention and Treatment 2016;23:12-3.
38. Li LK, Kuang WJ, Huang YF, et al. The effect of Astragalus injection on tumor suppression and immune function in H22 liver cancer mice. Sichuan Journal of Physiological Sciences 2011;33:14-6.
39. Lin XE, Song GC. Research on the effects of Astragalus polysaccharide injections on the expression of Ki-67, HIF-1α, and VEGF in HepG2-bearing mice during chemotherapy. Journal of North Pharmacy 2020;17:71-2.
40. Li LK, Kuang WJ, Huang YF, et al. Anti-tumor effects of Astragalus on hepatocellular carcinoma in vivo. Indian J Pharmacol 2012;44:78-81. [Crossref] [PubMed]
41. Lai X, Xia W, Wei J, et al. Therapeutic Effect of Astragalus Polysaccharides on Hepatocellular Carcinoma H22-Bearing Mice. Dose Response 2017;15:1559325816685182. [Crossref] [PubMed]
42. Pu XY, Fan WB, Yu S, et al. Influence of polysaccharides from angelica and astragalus onH22hepatocarcinoma mice. J Chem Pharm Res 2014;6:507-1.
43. Dongshan P. Research on the Synergy and Detoxifying Effects of Astragalus Polysaccharides on Hepatocellular Carcinoma Chemotherapy and Its Mechanisms. 2020. Available online: https://link.cnki.net/doi/10.27877/d.cnki.gmdjm.2020.000002
44. Qu X, Gao H, Zhai J, et al. Astragaloside IV enhances cisplatin chemosensitivity in hepatocellular carcinoma by suppressing MRP2. Eur J Pharm Sci 2020;148:105325. [Crossref] [PubMed]
45. Tang D, Zhang S, Shi X, et al. Combination of Astragali Polysaccharide and Curcumin Improves the Morphological Structure of Tumor Vessels and Induces Tumor Vascular Normalization to Inhibit the Growth of Hepatocellular Carcinoma. Integr Cancer Ther 2019;18:1534735418824408. [Crossref] [PubMed]
46. Wu JY, Shen Q, Jin XJ, et al. Inhibition of ascites tumor H22 in BALB/C mice by Astragalus membranaceus saponins and its mechanisms. Chinese Journal of Pharmacovigilance 2016;13:138-5.
47. Yan CL, Luo YL, An FY, et al. The inhibitory effect of total saponins from Astragalus membranaceus combined with cisplatin on STAT3-mediated inflammatory signaling pathways in H22 tumor-bearing mice. Chinese Journal of Biochemistry and Molecular Biology 2020;36:1359-6.
48. Yang B, Xiao B, Sun T. Antitumor and immunomodulatory activity of Astragalus membranaceus polysaccharides in H22 tumor-bearing mice. Int J Biol Macromol 2013;62:287-90. [Crossref] [PubMed]
49. Zhang S, Tang D, Zang W, et al. Synergistic Inhibitory Effect of Traditional Chinese Medicine Astragaloside IV and Curcumin on Tumor Growth and Angiogenesis in an Orthotopic Nude-Mouse Model of Human Hepatocellular Carcinoma. Anticancer Res 2017;37:465-73. [Crossref] [PubMed]
50. Zhang XF. The effect of total glycosides of Astragalus on the growth of diethylnitrosamine-induced hepatocellular carcinoma in rats. Journal of Zhengzhou University 2014;49:719-2. (Medical Sciences Edition).
51. Zeng PH, Gao WH, Pan MQ, et al. Effect of Prescription of Strengthening Qi and Clearing Poison and Toxin on Expressions of MVD,HIF1a and VEGF/KDR from Hepg2 Hepatocellular Carcinoma Nude Mice Transplantation Tumor. Chinese Archives Of Traditional Chinese Medicine 2014;32:1563-5.
52. Zeng P, Gao W, Pan M, et al. Effect of Yiqi Huayu Jiedu Formula and Its Separates on Nude Mice Transplantation Tumor of HepG2 Cells and Expression of HIF1a and E-cad. Liaoning Journal Of Traditional Chinese Medicine 2014;41:2004-6.
53. Zeng P, Gao W, Pan M, et al. Effects of the Qi-tonifying, blood-activating, and detoxifying formula formula and its single-component drugs on cancer-associated factor of hemangiomimicry in human hepatocellular carcinoma HepG2 xenografts in nude mice. Chinese Journal of Information on Traditional Chinese Medicine 2015;22:55-9.
54. He L, Xu K, Niu L, et al. Astragalus polysaccharide (APS) attenuated PD-L1-mediated immunosuppression via the miR-133a-3p/MSN axis in HCC. Pharm Biol 2022;60:1710-20. [Crossref] [PubMed]
55. Lin YT, Song GC. Experimental study on the antitumor effects and immunomodulatory properties of Scutellaria barbata polysaccharides. Information on Traditional Chinese Medicine 2022;39:51-4.
56. Lin ZC, Zhou HR, Zhong YZ, et al. The influence of Astragalus polysaccharide on the antitumor efficacy of PD-1 inhibitors in H22 tumor-bearing mice. Chinese Journal of Traditional Medical Science and Technology 2024;31:217-2.
57. Li C, Pan XY, Ma M, et al. Astragalus polysacharin inhibits hepatocellular carcinoma-like phenotypes in a murine HCC model through repression of M2 polarization of tumour-associated macrophages. Pharm Biol 2021;59:1533-9. [Crossref] [PubMed]
58. Li M, Duan F, Pan Z, et al. Astragalus Polysaccharide Promotes Doxorubicin-Induced Apoptosis by Reducing O-GlcNAcylation in Hepatocellular Carcinoma. Cells 2023;12:866. [Crossref] [PubMed]
59. Min L, Wang H, Qi H. Astragaloside IV inhibits the progression of liver cancer by modulating macrophage polarization through the TLR4/NF-κB/STAT3 signaling pathway. Am J Transl Res 2022;14:1551-66.
60. Ma Y, Li Y, Wu T, et al. Astragaloside IV Attenuates Programmed Death-Ligand 1-Mediated Immunosuppression during Liver Cancer Development via the miR-135b-5p/CNDP1 Axis. Cancers (Basel) 2023;15:5048. [Crossref] [PubMed]
61. Sun M, Wang J, Shi Z, et al. Effects of apatinib mesylate combined with astragaloside IV on proliferation,migration and invasion of primary hepatocellular carcinoma cells. Progress of Anatomical Sciences 2023;29:55-9+63.
62. Zeng PH, Gao WH, Pan MQ, et al. Effects of the Qi-tonifying, blood-activating, and detoxifying Chinese herbal formula on the proliferation of human liver cancer HepG2 cells and the growth of transplanted tumors in nude mice. Journal of Modern Oncology 2014.
63. Zhang Q, Li J, Zhang Q, et al. A Comparative Study on Anti-Cancer Effects of Astragalus membranaceus, Sophora flavescens and Gleditsia sinensis. Information on Traditional Chinese Medicine 2020;37:48-4.
64. Zhu Y, Lu F. Astragaloside IV inhibits cell viability and glycolysis of hepatocellular carcinoma by regulating KAT2A-mediated succinylation of PGAM1. BMC Cancer 2024;24:682. [Crossref] [PubMed]
65. Varadhan R, Wang SJ. Standardization for subgroup analysis in randomized controlled trials. J Biopharm Stat 2014;24:154-67. [Crossref] [PubMed]
66. Yang Y, Zheng B, Han Q, et al. Targeting blockage of STAT3 inhibits hepatitis B virus-related hepatocellular carcinoma. Cancer Biol Ther 2016;17:449-56. [Crossref] [PubMed]
67. Tian JH, Liu WD, Zhang ZY, et al. Influence of miR-520e-mediated MAPK signalling pathway on HBV replication and regulation of hepatocellular carcinoma cells via targeting EphA2. J Viral Hepat 2019;26:496-505. [Crossref] [PubMed]
68. Geng M, Xin X, Bi LQ, et al. Molecular mechanism of hepatitis B virus X protein function in hepatocarcinogenesis. World J Gastroenterol 2015;21:10732-8. [Crossref] [PubMed]
69. Huang Z, Zhou L, Duan J, et al. Oxidative Stress Promotes Liver Cancer Metastasis via RNF25-Mediated E-Cadherin Protein Degradation. Adv Sci (Weinh) 2024;11:e2306929. [Crossref] [PubMed]
70. Liu MX, Jin L, Sun SJ, et al. Metabolic reprogramming by PCK1 promotes TCA cataplerosis, oxidative stress and apoptosis in liver cancer cells and suppresses hepatocellular carcinoma. Oncogene 2018;37:1637-53. [Crossref] [PubMed]
71. Banerjee P, Gaddam N, Chandler V, et al. Oxidative Stress-Induced Liver Damage and Remodeling of the Liver Vasculature. Am J Pathol 2023;193:1400-14. [Crossref] [PubMed]
72. Kim DH, Sung M, Park MS, et al. Galectin 3-binding protein (LGALS3BP) depletion attenuates hepatic fibrosis by reducing transforming growth factor-β1 (TGF-β1) availability and inhibits hepatocarcinogenesis. Cancer Commun (Lond) 2024;44:1106-29. [Crossref] [PubMed]
73. Zhao X, Liu R, Zhou H, et al. Astragaloside IV represses hepatocellular carcinoma progression by modulating HMGB1-ferroptosis axis. Discov Oncol 2025;16:1454. [Crossref] [PubMed]
74. Wang C, Shi CH, Bai HY, et al. Astragali radix - Curcumae rhizoma herb pair suppresses hepatocellular carcinoma through EGFR/AKT/mTOR pathway and induces lipid peroxidation-related ferroptosis via HIF-1α/HO-1/GPX4 axis. J Ethnopharmacol 2025;348:119912. [Crossref] [PubMed]
75. Seifert L, Deutsch M, Alothman S, et al. Dectin-1 Regulates Hepatic Fibrosis and Hepatocarcinogenesis by Suppressing TLR4 Signaling Pathways. Cell Rep 2015;13:1909-21. [Crossref] [PubMed]
76. Gu S, Jiang M, Zhang B. Microcystin-LR in Primary Liver Cancers: An Overview. Toxins (Basel) 2022;14:715. [Crossref] [PubMed]
77. Pallozzi M, De Gaetano V, Di Tommaso N, et al. Role of Gut Microbial Metabolites in the Pathogenesis of Primary Liver Cancers. Nutrients 2024;16:2372. [Crossref] [PubMed]