Gastric cancer is the third leading cause of cancer mortality worldwide. Its development follows the classic pathological evolution of the Correa cascade reaction. Normal gastric mucosa successively undergoes chronic non-atrophic gastritis, chronic atrophic gastritis (CAG), intestinal metaplasia (IM), dysplasia (Dys), and finally progresses to invasive gastric cancer [1]. In this review, “CAG” refers to the chronic atrophic gastritis stage (glandular atrophy of the gastric mucosa); “IM” denotes intestinal metaplasia and “Dys” denotes dysplasia; “PLGC” (precancerous lesions of gastric cancer) is used specifically for the IM and Dys stages, which are regarded as the key reversible nodes in the transformation from gastritis to gastric cancer. Established invasive gastric cancer is a terminal-stage disease and is explicitly excluded from the disease scope to which the “reversal of PLGC” conclusions in this review apply. Evidence derived from gastric cancer cell lines or tumor-bearing mouse models is used only as mechanistic background and is clearly distinguished from direct evidence obtained in PLGC/PLGC-model systems throughout the text. Studies have shown that the prevalence of CAG can reach 20% - 30% in specific high-risk populations globally, and the risk of developing gastric cancer in patients with incomplete IM is about 13.4 times higher than that in patients with complete IM [2][3]. Therefore, clarifying the pathogenesis of PLGC and finding effective treatment strategies to block or even reverse its malignant transformation are of great significance for the secondary prevention of gastric cancer.

In recent years, traditional Chinese medicine (TCM) has demonstrated unique advantages in the prevention and treatment of PLGC. The comprehensive intervention characteristics of TCM, with multiple components, multiple targets, and multiple pathways, are highly consistent with the complex pathological mechanisms of PLGC. Multiple studies have confirmed that Chinese herbal compounds can significantly improve the clinical symptoms of CAG patients and reverse glandular atrophy and intestinal metaplasia at the histopathological level [4][5]. This provides a solid practical foundation for TCM intervention in PLGC and implies that its mechanism of action may be closely related to the regulation of multiple signaling pathways.

FQSG has been widely used clinically in the treatment of CAG. Preliminary clinical observations have confirmed that this formula can reduce the infiltration of inflammatory cells, increase the thickness of the gastric mucosa, and reverse glandular atrophy and intestinal metaplasia. However, the exact molecular mechanism by which it exerts its anti-PLGC effect remains to be systematically elucidated. Modern pharmacological studies have shown that PLGC involves the abnormal activation and cross - talk of multiple signaling pathways such as Wnt/β-catenin, PI3K/Akt, NF-κB, JAK/STAT and Hippo [6]-[8]. Therefore, analyzing the molecular mechanism of FQSG against PLGC based on the cross-regulatory network of multiple signaling pathways has important academic and application values. This paper aims to systematically review the relevant research progress in recent years and explore the molecular mechanism by which it exerts its therapeutic effect through the regulation of multiple key signaling pathways and their cross-networks, in order to provide a theoretical basis for in-depth research and clinical application.

To avoid overclaiming, this review applies an explicit three-tier evidence classification. i) Direct evidence: studies in which FQSG itself was tested in PLGC, CAG, or PLGC-related models; only these data are used to support definitive conclusions about FQSG mechanisms. ii) Indirect/analogous evidence: findings from other herbal formulas or single compounds in PLGC or CAG models; these are explicitly labeled as “analogous interventions suggest… but have not been directly validated in FQSG” and are treated as hypotheses rather than established FQSG mechanisms. iii) Background evidence: studies in gastric cancer cell lines, tumor-bearing mice, or human gastric cancer tissues; these serve only to establish the relevance of a given pathway to gastric carcinogenesis and are never used as evidence that FQSG reverses PLGC.

FQSG are an adaptation of the traditional Qifang Stomachache-relieving Granules, designed to “invigorate the spleen and replenish qi, clear heat and resolve dampness, and promote blood circulation to remove stasis.” The constituents and dosages of FQSG are as follows: Red Ginseng 10 g, Poria Cocos 10 g, Largehead Atractylodes Rhizome 10 g, Honey-fried Licorice Root 9 g, Coptis Rhizome 6 g, Aucklandia Root 6 g, Evodia Fruit 3 g, Immature Bitter Orange 6 g, Pinellia Tuber 10 g, Astragalus Root 30 g, Solomonseal Rhizome 6 g, Snakegourd Peel 20 g, Snakegourd Root 10 g, White Peony Root 30 g, Salvia Miltiorrhiza Root 10 g, Stir-fried Chicken’s Gizzard-membrane 9 g, Burreed Tuber 15 g, Zedoary Rhizome 15 g, Spreading Hedyotis Herb 20 g. The formula primarily uses Codonopsis Radix and Atractylodis Macrocephalae Rhizoma as the main components. Supporting ingredients like Salviae Miltiorrhizae Radix and Rhizoma and Curcumae Zedoariae Rhizoma enhance blood circulation, aiming to alleviate blood stasis caused by qi stagnation and damp obstruction, thereby improving gastric mucosal microcirculation. Adjuvant ingredients such as Citri Reticulatae Pericarpium and Pinelliae Rhizoma help regulate qi, harmonize the stomach, relieve nausea and vomiting, and balance the formula’s properties. This comprehensive approach addresses the complex pathology of chronic atrophic gastritis (CAG), characterized by “deficiency in origin and excess in superficiality.”

Modern pharmacological research highlights astragaloside IV, a key active component from Astragali Radix, for its anti-inflammatory, antioxidant, and immunomodulatory properties. It mitigates oxidative damage to the gastric mucosa by modulating the ferroptosis-related signaling pathway [9]. Berberine from Coptidis Rhizoma and dehydroevodiamine (DHE) from Evodiae Fructus form a pair that clears heat and dries dampness. DHE has been shown to improve gastric mucosal damage in CAG rats induced by N-methyl-N’-nitro-N-nitrosoguanidine (MNNG) and inhibit GES-1 cell migration and invasion by targeting the HIF-1α/VEGF angiogenesis pathway [10]. Tanshinone IIA from Salviae Miltiorrhizae Radix et Rhizoma also regulates ferroptosis-related pathways [9]. Network pharmacological analysis identifies naringenin, luteolin, and quercetin as crucial components that work synergistically through multiple targets [11]. It should be noted that most of the monomer-level mechanistic data in this paragraph derive from active compounds tested individually or from network pharmacology predictions, rather than from FQSG as a whole administered in PLGC models; they are therefore treated as pharmacological hypotheses regarding FQSG’s potential mechanism rather than established findings.

These active components interact within a multi-target, multi-pathway synergistic network. Astragaloside IV and tanshinone IIA inhibit cell proliferation and induce apoptosis via the PI3K/Akt signaling pathway. The combination of berberine and dehydroevodiamine reduces inflammation and gastric mucosal damage in H. pylori-induced CAG rat models by inhibiting the JMJD2B/COX-2/VEGF axis and the HMGB1/NF-κB signaling pathway [12]. The “monarch, minister, assistant, and conductant” principle is evident at the molecular level, where different active components synergistically regulate various nodes of the same signaling pathway or intervene in multiple parallel pathways. This multi-component, multi-target synergistic mechanism is considered a plausible advantage of this formula, although direct confirmation in intact FQSG-treated PLGC models remains to be established.

The pathological progression of PLGC is characterized by distinct stage-specific features. In the context of CAG, pathological changes include atrophy of the gastric mucosal epithelium and glands, marked by a reduction in gland numbers, thinning of the mucosal layer, and often thickening of the muscularis mucosa. This is sometimes accompanied by pyloric or intestinal gland metaplasia, and dysplasia may be observed in certain cases. The PLGC rat model, induced by N-methyl-N’-nitro-N-nitrosoguanidine, similarly exhibits significant pathological changes such as atrophy, intestinal epithelial metaplasia, and dysplasia [13]. As the lesion advances, the gastric mucosal epithelium is replaced by intestinal-type cells, forming intestinal metaplasia (IM), a common precancerous lesion of intestinal-type gastric cancer. Pathological features of IM include the presence of intestinal-type epithelial cells like goblet, Paneth, and absorptive cells in the gastric mucosa, indicating abnormal cell differentiation [14]-[16]. IM and dysplasia (Dys) have also been successfully induced in Atp4a-/- mice and the gastrojejunostomy mouse model. Dysplasia is a stage with higher malignant potential in PLGC, characterized by abnormal cell morphology and tissue structure. Histologically, gastric dysplasia is often accompanied by an inflammatory background (60.42%) and intestinal epithelial metaplasia (64.58%), with a blurred lesion boundary and retention of the maturation gradient (97.72%). The nuclei are typically round (97.92%), with a high nuclear-cytoplasmic ratio, though cell polarity is usually maintained (89.58%). Identifying dysplasia, compared to low-grade intraepithelial neoplasia (LGD) and high-grade intraepithelial neoplasia (HGD), requires immunohistochemical analysis. In Dys, MUC5AC/MUC6 expression retains the maturation gradient but is significantly reduced or absent in LGD/HGD. P53 is mainly expressed in its mutant form in HGD (66.67%), while in Dys and LGD, it is in the wild-type form (100%/93.33%). The expression pattern of Ki67 shifts from a “gastric pit pattern” in Dys to a “polar pattern” in LGD and a “diffuse pattern” in HGD. This combined analysis of cell morphology and molecular markers is crucial for the objective diagnosis and differentiation of various stages of gastric precancerous lesions [17][18].

The therapeutic mechanism of FQSG in treating PLGC involves a complex cross-regulatory network rather than a single linear signaling pathway. Transcriptomic analysis reveals numerous differentially expressed genes (DEGs) in the gastric tissues of PLGC rats following treatment with these granules, with significant enrichment in the PI3K-Akt signaling pathway [29]-[31]. This pathway is crucial for regulating cell survival and proliferation, as its activation leads to the phosphorylation and inhibition of downstream transcription factors like FoxO3a, promoting proliferation and inhibiting apoptosis. Additionally, the PI3K-Akt pathway interacts extensively with the Raf-1/MEK/ERK pathway. Raf-1, a pivotal node in the RAS-RAF-MEK-ERK signaling cascade, is significantly down-regulated by FQSG in a chronic atrophic gastritis rat model [46]. Activation of Raf-1 not only advances the ERK signal to promote the cell cycle but also co-regulates apoptosis and autophagy by phosphorylating Bcl-2 family members or interacting with Akt. Thus, the granules may inhibit both the PI3K-Akt and Raf-1/ERK pathways, synergistically blocking the malignant transformation of PLGC.

Moreover, cell-cycle regulation is central to PLGC progression. FQSG significantly down-regulate CDK2 and CDK4 protein expression in the gastric mucosa of PLGC rats [47]. These proteins are key drivers of the G1/S checkpoint in the cell cycle, co-regulated by the PI3K-Akt and Raf-1/ERK pathways. The ERK signal enhances Cyclin D1 expression and CDK4/6 activation, while Akt alleviates CDK2 inhibition by suppressing CDK inhibitors like p21 and p27. The down-regulation of CDK2/4 by the granules likely results from the downstream cascade effect following the inhibition of the PI3K-Akt and Raf-1/ERK pathways. Concurrently, the granules reduce the expression of p53 and Ki-67. It should be emphasized that the expression level of p53 must be distinguished from its functional status. Wild-type p53 has a short half-life and is barely detectable under normal conditions; the p53 protein accumulation detected in dysplasia and gastric cancer typically reflects the abnormal accumulation of mutant p53 (which loses its tumor-suppressive function and may even acquire oncogenic properties). Reported positive rates rise along the progression from intestinal metaplasia (approximately 12.1%) to dysplasia (approximately 34.6%) and gastric cancer (approximately 61.5%), with p53 gene mutation rates increasing from approximately 31.3% in dysplasia to approximately 60.1% in gastric cancer. Therefore, the decreased p53 protein positivity observed after FQSG intervention should be interpreted as inhibition of pathological p53 abnormal accumulation rather than restoration of normal wild-type p53 tumor-suppressive function. Since no mutation-sequencing or functional data are provided in this review, the downregulation of p53 should be reported as an observational finding rather than a mechanistic marker of curative reversal. Ki-67, in contrast, serves as a reliable marker of cell proliferation.

The comprehensive therapeutic effect of FQSG arises from precise intervention in multiple key nodes. The PI3K-Akt and Raf-1/ERK pathways are core hubs, influencing each other through shared downstream effectors (mTOR, GSK-3β) and negative-feedback mechanisms (Akt phosphorylates Raf-1 to inhibit its activity). By inhibiting both pathways, the granules prevent compensatory activation, effectively blocking pro-cancer signals. Additionally, regulation of CDK2/4 and p53 directly impacts cell cycle and DNA damage repair modules. This multi-level network regulation may explain why the granules outperform single-target drugs in improving gastric mucosa pathology and reducing inflammatory factors (IL-6, TNF-α) in PLGC rats [31]. Identifying key nodes like PI3K, Akt, Raf-1, and CDK2/4 holds significant clinical importance.

Leveraging its formula’s strengths in enhancing spleen function, replenishing qi, clearing heat, removing dampness, and promoting blood circulation, FQSG regulates core signaling pathways such as Wnt/β-catenin, PI3K/Akt, NF-κB, MAPK, and TGF-β. This creates a multidimensional cross-regulatory network that effectively inhibits chronic gastric mucosa inflammation, blocks abnormal cell proliferation, prevents epithelial-mesenchymal transition, and stabilizes gastric mucosal cell homeostasis, reversing pathological changes like glandular atrophy, intestinal metaplasia, and dysplasia in PLGC. Compared to single-target drugs, the granules’ multi-pathway, network-based regulation aligns better with PLGC’s complex pathology. Future research should deepen understanding of this multi-pathway cross-network mechanism and enhance clinical translation to clarify its scientific basis.

Several limitations regarding the mechanistic evidence should be noted. For several pathways, including Wnt/β-catenin and Notch, the evidence is either derived from indirect/background studies (e.g., gastric cancer tissue observations, or other Chinese herbal formulas in PLGC models) rather than from FQSG-treated PLGC models, or no FQSG-specific data are yet available. Regarding p53, the observed downregulation is reported as an observational finding rather than direct mechanistic evidence, owing to the absence of mutation-status and functional assays. Future studies should therefore directly validate these pathway modulations in FQSG-treated PLGC models, ideally incorporating p53 functional/mutation analyses, to distinguish hypothesis-driven mechanisms from confirmed FQSG actions. Methodological caveat: owing to the absence of p53 mutation-status and functional assays in the cited studies, the observed p53 downregulation is treated in this review as an observational finding rather than direct mechanistic evidence of restored wild-type p53 tumor-suppressive function. This work offers new theoretical and practical support for preventing and treating precancerous gastric lesions and inhibiting inflammation-cancer transformation with traditional Chinese medicine.

Project for Improving the Basic Scientific Research Abilities of Young and Middle-aged Teachers in Guangxi Universities (2024KY0320).

*First author.

^#Corresponding author.