The FYD consists of nine different herbs: Actindia arguta (Siebold & Zucc.) Planch. ex Miq. [Actinidiaceae; Actindia arguta root] 25 g, S. barbata D.Don [Lamiaceae; Scutellaria barbata whole plant] 20 g, Scleromitrion diffusum (Willd.) R.J.Wang [Rubiaceae; Scleromitrion diffusum whole plant] 40 g, Iphigenia indica (L.) A.Gray ex Kunth [Colchicaceae; Iphigenia indica pseudobulb] 4 g, Salvia miltiorrhiza Bunge [Lamiaceae; Salviae miltiorrhizae radix et rhizome] 20 g, G. lucidum (Curtis) P. Karst. 15 g, Panax ginseng C.A.Mey. [Araliaceae; Panax ginseng radix et rhizome] 12 g, Lycium chinense Mill. [Solanaceae; Lycium chinense fruit] 12 g, and Astragalus mongholicus Bunge [Fabaceae; Astragalus mongholicus root] 20 g. All herbs were purchased from Jiangxi Kangqingtang Herbal decoction Pieces Co., Ltd. (Zhangshu, China). FYD standard decoction was prepared according to “Management standard of traditional Chinese medicine (TCM) decoction in medical institutions” (Wang et al., 2019). The water-mixed raw materials were soak for 30 min, then decocted for another 45 min and filtered for collection of the filtrate. The residue was decocted again and filtered twice. The filtrates were then freeze-dried to obtain FYD powder.

Compound identification of FYD, cell culturing, cell viability assay, cell cycle assay, cell apoptosis assay, hetero-adhesion assay, migration assay, invasion assay, flow cytometry, western blot, RNA extraction and qRT-PCR assay, in vivo pulmonary metastasis assays are described in the online Supplementary Material. The primer sequences for qPCR were presented in Supplementary Table S1.

Specific pathogen-free female BALB/c mice (20–22 g, 6–8 weeks old) were procured from the Shanghai SLAC Laboratory Animal Co., Ltd (Lu et al., 2019). The mice were housed in a controlled environment with no exposure to pathogens and provided ad libitum access to pelleted food and water. All animal experiments adhered to the guidelines outlined in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council, 1996). All animal experimental procedures were thoroughly evaluated and approved by the Laboratory Animal Ethics Committee of Minjiang University.

The results were expressed as means and standard deviations. Statistical analyses were conducted using the SPSS statistical software package. Differences in experimental data were assessed through analysis of variance (ANOVA), with Dunnett’s method employed for adjusting the ANOVA for multiple comparisons. A significance level of p < 0.05 was considered statistically significant, while a significance level of p < 0.01 was considered extremely statistically significant.

UPLC-HRMS analysis revealed that FYD comprises a total of 1430 chemical components, which can be classified and identified into 107 categories (Supplementary Table S2). The categories of the top 6 compounds and their subclasses were shown in Supplementary Table S3, and the proportion of the number of the top 6 compounds and their subclasses was shown in Supplementary Figure S1. The peaks exhibiting higher abundance in the positive and negative ion base peak chromatograms (BPC) underwent confirmation of their peak shape and secondary spectral examination prior to being assigned with numerical labels. Sequential numbering was applied to both the positive and negative ion chromatograms (Figures 1A,B). The 40 peaks identified in the positive and negative ion BPC diagram for FYD composition analysis mainly comprise carboxylic acids, purine nucleosides, coumarins, flavonoids, organic oxides, hydroxy acids, cinnamic acids, lipids and other classes of compounds. The detailed information of the compounds was shown in Supplementary Table S4.

The cytotoxicity of FYD was evaluated in CNE1 cells, revealing a concentration-dependent reduction in cell viability with IC50 and IC10 (Lu et al., 2019) values of 4.695 mg/mL and 4 mg/mL, respectively (Supplementary Figure S2). We also examined the effects of FYD on the proliferative activity of normal cell lines human nasal epithelial cells HNEPC and human umbilical vein endothelial cells (HUVECs), with IC50 values of 5.218 mg/mL and 4.789 mg/mL, respectively (Supplementary Figures S3, S4). Furthermore, based on the results of cell cycle and apoptosis assays, it was observed that medication concentrations equal to or below 4 mg/mL did not exhibit any discernible effect on CNE1 cells, and apoptosis was not produced (Figures 1C–F). These findings suggest that FYD possesses weak cytotoxicity at concentrations less than 4 mg/mL. Therefore, in order to elucidate the mechanism underlying the cancer metastasis preventive effect of FYD, subsequent experiments were conducted using FYD concentrations at or below 4 mg/mL, aiming to demonstrate its inhibitory effect on CTCs adhesion and immune enhancement rather than cytotoxicity-mediated CTCs elimination.

The adhesion of CTCs to endothelial cells is a critical step in the metastasis of cancer cells to distant sites (Stott et al., 2010). Here, we investigated the impact of FYD on the heterogeneous adherence between TNF-α-induced cancer cell CNE1 and HUVECs. The results of the adhesion assay between CNE1 cells and HUVECs showed that FYD effectively inhibited adhesion with a concentration-dependent trend. In the presence of FYD doses of 0.5, 1.0, 2.0, and 4.0 mg/mL, the adhesion rates between CNE1 and HUVECs were observed to be 77.1% ± 12.8%, 30.5% ± 1.8%, 21.5% ± 3.7%, and 7.4% ± 1.1%, respectively, as compared to the control group (induced by TNF-α) (Figures 2A,B). The present study demonstrated that FYD effectively suppressed TNF-α-induced adhesion of CNE1 to HUVECs at non-toxic concentrations.

The pro-inflammatory cytokine TNF-α plays a pivotal role in cancer metastasis by inducing the expression of CAMs on HUVECs (Lian et al., 2016). We employed flow cytometry to assess the effect of FYD on the expression of E-selectin (CD62E), ICAM-1 (CD54), and VCAM-1 (CD106) in HUVECs. The experimental results showed that FYD effectively inhibited the expression of VCAM-1, ICAM-1, and E-selectin in HUVECs with a concentration-dependent trend (Figures 2C,D). This finding prompted us to further investigate the molecular mechanism underlying FYD mediated disruption of CNE1 cells adhesion to endothelial cells in subsequent investigations.

The NF-κB signaling pathway has been extensively demonstrated to play a pivotal role in the regulation of expression of CAMs (Jayakumar et al., 2014; Zeng et al., 2021). Here, we employed western blot analysis to investigate the effect of FYD on the expression of CAMs in HUVECs via modulation of the NF-κB signaling pathway. As shown in Figures 3A–F, the NF-кB signaling pathway is activated by TNF-α through upregulation of NF-кB p65 expression and phosphorylation of p-IкBα (Ser 32), p-IKKα/β (Ser 176/180) and p-NF-кB p65 (Ser 536). Treatment with FYD in HUVECs resulted in a significant concentration-dependent reduction in the total protein level of IкBα and NF-кB p65, as well as the phosphorylation levels of p-IкBα (Ser 32), p-IKKα/β (Ser 176/180) and p-NF-кB p65 (Ser 536), compared to the TNF-α stimulated HUVECs. Additionally, the p-IKK/IKK ratio exhibited a significant reduction in comparison to HUVECs stimulated with TNF-α. We performed additional analysis to investigate the impact of FYD on NF-кB p65 nuclear translocation using immunofluorescence staining. The results revealed a significant dose-dependent inhibition of NF-кB p65 nuclear translocation in HUVECs by FYD (Figures 3G,H). These findings suggest that FYD exerts inhibitory effects on the NF-κB signaling pathway, leading to the degradation of NF-кB p65 through blockade of its nuclear translocation. Consequently, the downregulation of CAMs expression on HUVECs by FYD is closely associated with modulation of the NF-κB signaling pathway.

Another crucial step in the metastatic process involves the transmigration of CTCs across endothelial cells lining blood vessels, with cancer cell aggressiveness playing a pivotal role in this intricate mechanism (Wang et al., 2022). Here, we investigated the effect of FYD on the migratory and invasive behavior of CNE1 cells in vitro. The wound-healing assays revealed a concentration-dependent inhibition of CNE1 cell migration by FYD, with noticeable effects observed at concentrations of 3 mg/mL and 4 mg/mL after 48 h (Figures 4A–C). Furthermore, transwell assays demonstrated that FYD effectively suppressed CNE1 cell invasion in a concentration-dependent manner, with the group treated with 4 mg/mL showing only 6% invasion compared to the control group (Figures 4D,E). These results indicate that FYD exerts a significant inhibitory effect on the migration and invasion of CNE1 cells.

The EMT process is strongly associated with cancer cell migration and invasion (Cheng et al., 2018; Liu et al., 2022). Therefore, we performed western blotting analysis to determine the effect of FYD on expression of EMT-related proteins in CNE1 cells. The results are shown in Figures 5A–C, FYD produced concentration-dependent inhibition of ZEB1, N-cadherin, β-catenin and MMP2 proteins expression, compared with the control group. We also performed qRT-PCR analysis to investigate the mRNA levels of EMT transcription-related factors, and found that the transcription levels of transcription factors ZEB1, FN1, and SNAIL2 was significantly reduced by FYD (Supplementary Figure S5).

The phosphatidylinositol-3-kinase (PI3K)/AKT signaling pathway participates in migration process of metastatic cancer cells, including the regulation of cytoskeleton-remodeling proteins and EMT–activating proteins that specifically regulate cell motility (Xue et al., 2012; Xue and Hemmings, 2013). We investigated the effect of FYD on the PI3K/AKT signaling pathway through western blot assay. The results demonstrated a significant concentration-dependent inhibition of p-PI3K (Tyr 458), AKT 1/2/3, and p-AKT 1/2/3 (AKT1-Tyr 315/AKT 2-Tyr 316/AKT 3-Tyr 312) expression by FYD, as compared to the control group (Figures 5D–F). Additionally, the p-PI3K/PI3K ratio exhibited a significant reduction. These findings suggest that FYD may inhibit the PI3K/AKT signaling pathway by inducing phosphorylation and total protein degradation. Moreover, the FAK signaling axis plays a pivotal role in facilitating cancer cell migration and invasion (Zhang et al., 2022). According to our findings, FYD exhibited a dose-dependent inhibition on the expression of FAK, p-FAK (Tyr 397), and p-PTK2 (Tyr 576/577) (Figures 5G,H). As a result of these findings, FYD suppresses the invasion and migration of cancer cell by regulating the EMT, PI3K/AKT, and FAK signaling cascades.

Due to the challenges in modeling nasopharyngeal carcinoma metastasis in immunocompetent syngeneic mice, the mechanisms of lung metastasis based on CTCs are similar for both breast cancer and nasopharyngeal cancer. In addition, in vitro CCK8 and transwell assays also showed that FYD could concentration-dependently inhibit the survival and invasion of mouse breast cancer 4T1 cells (Figures 6A–C), which was consistent with the results obtained in human nasopharyngeal carcinoma CNE1 cells (Figure 4D). Therefore, in this study, we utilized 4T1 breast cancer cells to examine the potential preventive effect of FYD on pulmonary metastasis in vivo. In this experiment, BALB/c mice were pretreated with FYD (0, 400, 800, and 1600 mg/kg) for a duration of 3 days, and then the mice received an intravenous injection of 4T1 cells via the tail vein. The doses administered to BALB/c mice were converted from clinical doses and in vitro 4T1 cell experiments (Bae et al., 2015; Nair et al., 2018). The administration of FYD continued to observe lung metastasis and survival rates. The results as shown in Figures 6D,E, revealed that FYD at doses of 800 mg/kg and 1600 mg/kg significantly reduced the formation of lung metastatic nodules induced by 4T1 cells. Histological hematoxylin and eosin (H&E) staining in Figure 6F demonstrated that the FYD-treated group had significantly fewer lung metastases compared to the control group. Moreover, FYD treatment extended the lifespan of mice with lung metastases, as depicted in Figure 6G. Immunohistochemical staining also showed that the expression of Ki67 was reduced in the lung metastasis tissue after FYD treatment, indicating that FYD treatment led to a weakened proliferation ability of 4T1 cells in lung tissue (Supplementary Figure S6). Furthermore, we further evaluated the effect of FYD for the PI3K/AKT and NF-κB signaling pathways in the lungs of mice through immunohistochemical staining. Interestingly, lung tumor metastases exhibited higher expression levels of PI3K, AKT and NF-κB p65 compared to healthy lung tissues; however, their expression significantly decreased following FYD treatment (Supplementary Figure S7). These findings suggest that FYD also effectively inhibits the PI3K/AKT and NF-κB signaling pathways in vivo. Importantly, no noticeable side effects were observed in the mice after 3 weeks of FYD treatment. The FYD treatment did not induce any significant changes in the body weight of the mice, while histological examination using H&E staining revealed that the FYD-treated mice exhibited normal morphological characteristics in their heart, liver, spleen, and kidney tissues (Supplementary Figure S8). The in vivo data revealed that FYD, even at weakly toxic doses, exhibits a remarkable inhibitory effect on the pulmonary metastasis.

It has been proposed that enhancing the immune system’s function and boosting immunity can help prevent cancer metastasis in the body (Lian et al., 2019). To investigate the effect of FYD on cellular immune function, flow cytometry was employed to measure the proportion of cytotoxic T lymphocytes (Tc cells) and natural killer cells (NK cells) in the peripheral blood. As shown in Figures 6H,I, FYD treatment significantly augmented the proportion of Tc cells (CD45⁺CD3⁺CD8⁺ cells) and NK cells (CD45+CD49b+ cells) in the peripheral blood, approaching physiological levels. These findings suggest that FYD reduces cancer metastasis and strengthens the anti-cancer immune response by promoting the release of Tc and NK cells by the immune system.