With the assistance of an information specialist, a search strategy was developed and refined using Boolean operators. The following search terms were used in the PubMed database: (“Renal Cell Carcinoma” OR RCC) AND (Metastasis OR metastatic) AND (Stomach OR Gastric). The search process and reporting were conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. The review protocol was not pre-registered. The initial search yielded a total of 365 articles. After removing duplicates, records flagged by automation tools as non-compliant, and excluding others for various reasons, 362 articles were screened. Of these, 139 articles were excluded for not discussing stomach metastases separately, and 34 were non-English articles. Ultimately, 189 articles were identified for full-text retrieval, with seven unable to be obtained. Following further eligibility assessment, 182 reports were evaluated, with 2 excluded due to severe data incompleteness, resulting in a total of 180 studies included for analysis.

This study systematically searched and included relevant original studies, conference abstracts, and case reports published between 1 January 2000, and 31 December 2025. The included literature was required to be based on human renal cancer tissue specimens and include immunohistochemical analysis. The exclusion criteria were as follows: (1) studies involving only cell lines or in vitro models; (2) studies focusing primarily on hereditary renal cancer syndromes; (3) studies that only analyzed recurrent renal cell carcinoma or advanced patients without independent analysis of metastatic sites; (4) studies restricted to subjects with renal failure, dialysis, or kidney transplant recipients; (5) studies with incomplete data or where key information could not be extracted. Criteria were specified a priori. The PRISMA flowchart is in Figure 1.

Two independent reviewers (Luling Wei, Runchang Liu) screened titles/abstracts. Discrepancies were resolved by a third reviewer (Xiaoqian Wang).

As mentioned earlier, RCC is broadly classified into clear cell carcinoma, papillary RCC, chromophoric carcinoma, oncocytic carcinoma and collecting duct carcinoma (Xu et al., 2021; Vanderbruggen et al., 2022; B et al., 2024; Rizzo et al., 2023). Clear cell carcinoma is the most common pathological subtype of RCC, accounting for approximately 60%–85% of RCC (Xu et al., 2021). Its hallmark is the abundant accumulation of lipid droplets within tumor-cell cytoplasm, imparting a clear or pale-yellow appearance under light microscopy (Goswami et al., 2024). Papillary RCC, also termed as chromophilic cell carcinoma or Leber’s RCC, comprises 7%–14% of tumors and is composed of basophilic or eosinophilic cells arranged in papillary or tubular architectures (Linehan et al., 2016). Oncocytic carcinoma is characterized by large volume, presence of solid cystic components, bleeding, and necrosis, accounting for approximately 2%–4% of RCCs. The pathological feature of chromophobe cell carcinoma is that tumor cells are arranged in a papillary or tubular structure. And its incidence rate accounts for 12%–14% of RCC. Renal collecting duct cancer, arising from the epithelium of Bellini ducts, is rare (0.4%–2%) and is characterised by aggressive growth, early renal-vein invasion and rapid dissemination to regional lymph nodes, bone, lung and liver (B et al., 2024; Deng et al., 2023). The clinical manifestations of patients often include back pain, gross hematuria, and palpable abdominal masses (Hsieh et al., 2017).

Many patients with RCC only experience symptoms in the late stages, such as bone pain, worsening physical condition, or persistent coughing (Ljungberg et al., 2022). The typical triad of lower back pain, gross hematuria, and palpable abdominal masses is now rare and mostly associated with late-stage diseases and subtypes associated with poor prognosis (Ljungberg et al., 2019). Most tumors are detected incidentally on imaging. Paraneoplastic features such as hypercalcemia, pyrexia of unknown origin or polycythemia may also be present. Patients suspected of RCC should immediately undergo laboratory tests for serum creatinine, hemoglobin, white blood cell and platelet counts, lymphocyte to neutrophil ratio, lactate dehydrogenase, C-reactive protein, and serum corrected calcium, and other relevant symptom-based tests (Escudier et al., 2016). Imaging examinations play an important role in the staging, preoperative planning, and follow-up of RCC. The main imaging methods for diagnosing RCC include US, computed tomography, and magnetic resonance imaging. CT examination can detect RCC with small lesion diameter, manifested as abnormal masses on the kidney accompanied by extensive tumor infiltration into the renal parenchyma, as well as perirenal soft tissue retention (Sankineni et al., 2016). Ultrasound (US) can be used for the initial evaluation of indeterminate renal masses. It is a valuable and non-invasive first-line imaging tool, but its role is primarily for screening and triage rather than definitive diagnosis. RCC on ultrasound may appear as a solid mass that is hypoechoic or isoechoic relative to the renal parenchyma, or it may present as a complex cystic mass with internal septations, wall thickening, or enhancing solid components (nodules). These features typically correspond to Bosniak category IIF, III, or IV lesions and often necessitate further specialized renal mass protocol CT or MRI for more precise staging and definitive diagnosis (Sankineni et al., 2016; Krishna et al., 2020). Compared to the above two methods, MRI provides superior soft-tissue contrast, clearly delineating septa, solid components and intra-tumoral hemorrhage, and is reserved for problem-solving or surveillance (Krishna et al., 2020). However, in the process of rapid improvement, CT remains the preferred method for diagnosing RCC, while US has become the most commonly used technique due to its convenience, and magnetic resonance imaging has become an auxiliary way to monitor patients. Additionally, the diagnosis of RCC can also be achieved through renal tumor biopsy (Ljungberg et al., 2019). Coarse needle biopsy is superior to fine needle aspiration and is avoided in purely cystic lesions due to its low diagnostic rate. Histopathology obtained from biopsy or nephrectomy specimens supplies key prognostic data, including nuclear grading, sarcomatoid features, vascular infiltration, tumor necrosis, collection system and perirenal fat infiltration, as well as evaluation of LN status (Ljungberg et al., 2022). In addition, histopathological diagnosis of malignant tumors can be obtained through renal core biopsy or partial or radical nephrectomy specimens (Cohen et al., 2005). A confirmatory biopsy is therefore recommended before ablation in non-surgical candidates or before starting systemic therapy in metastatic disease.

For local RCC, surgery remains only curative treatment method, which can be divided into radical nephrectomy (RN) and partial nephrectomy (PN). Tumors with a diameter of less than 7 cm are preferentially treated with PN; whereas kidney tumors larger than 7 cm are managed by open RN as the standard approach. In metastatic RCC, resection of the metastatic lesion is still the appropriate local treatment method for most metastatic sites by default. In addition, drug-eluting bead transarterial chemoembolization (DEB-TACE), a technique that allows drugs to slowly release into tumor tissue after arterial embolization, has demonstrated excellent efficacy in RCC patients with hepatic and retroperitoneal lymph node metastasis (Qiu et al., 2024).

The metastatic process of RCC cells was found to be driven by the coordinated activation of the PI3K/AKT, Ras/MAPK, and Wnt/β-catenin pathways. Through these pathways, proliferation, survival, and migratory cues were provided, enabling tumor cells to withstand adverse microenvironments and to establish secondary growth at distant sites.

Angiogenesis is a key factor in tumor growth and metastasis, especially play a key role in RCC (Table 3). Tumor cells promote the formation of new blood vessels through various pathways, providing the necessary nutrients and oxygen for the tumor while also promoting the spread and metastasis of tumor cells (Yhee et al., 2012). Neovascularisation is orchestrated through the coordinated activation of the vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF) and mammalian target of rapamycin (mTOR) signaling axes, whose induction supplies nutrients and oxygen while facilitating tumor cell dissemination (Négrier and Raymond, 2012). The expression of various target genes of VEGFA and PDGF is upregulated by HIF (Hypoxia-inducible factors), which is a transcription factor that regulates genes related to oxygen sensing, angiogenesis, and cell growth, and also a target gene of the tumor suppressor gene VHL (von Hippel Lindau). VHL is a tumor suppressor gene that encodes pVHL. The main function of pVHL is to target and degrade HIF, especially HIF-1α and HIF-2α, under normal oxygen conditions (Strikic et al., 2023). Although VEGF signaling predominates, PI3K/AKT/mTOR hyper-activation provides an additional route for HIF-1α stabilization (Posadas et al., 2013). VEGF also integrates the activation of the HIF and AKT/mTOR signaling pathways in ccRCC. Polycystin-1 (PC1), a large multi-pass membrane protein with GPCR-like domains, activates the PI3K/AKT/mTOR pathway, upregulates VEGF, and promotes angiogenesis (Gargalionis et al., 2020).

The RUNX family—RUNX1 (Runt-related transcription factor 1), RUNX2 (Runt-related transcription factor 2), and RUNX3 (Runt-related transcription factor 3)—has been implicated in both normal development and oncogenesis, with RUNX3 functioning as a tumor-suppressor gene across multiple malignancies. RUNX3 expression has been shown to downregulate VEGF production and secretion in RCC cells, thereby attenuating cell migration, invasion, and angiogenesis (Chen et al., 2013). Ribosomal S6 protein kinase 4 (RSK4), a component of the MAPK axis, not only directly stimulates VEGF secretion, but also upregulates EPHA2 by phosphorylating RUNX1, further promoting the migration, invasion and angiogenesis of RCC cells (Ma et al., 2025). Mir-125a-3p can downregulate VEGF expression under the activation of PinX1 (PIN2/TRF1-interacting telomerase inhibitor 1), thereby inhibiting RCC angiogenesis and thus inhibiting RCC progression (Hou et al., 2019). CHIP (C-terminal Hsp70-interacting protein) is an E3 ubiquitin ligase that inhibits RCC cell migration, invasion, and angiogenesis by suppressing the VEGF-VEGFR2 pathway (Sun et al., 2015). RASAL2 is a unique RAS GTPase activating protein (RAS-GAP), which can activate GSK3β by reducing Ser9 phosphorylation, thereby reducing VEGFA expression (Hui et al., 2018). F-box only protein 22 (FBXO22), a substrate receptor of the SKP1-Cullin 1-F-box protein (SCF) E3 ubiquitin ligase, has been found to inhibit VEGF-mediated angiogenesis (Guo et al., 2019). Maslinic acid has been observed to inhibit VEGF expression and to inhibit cell proliferation (Thakor et al., 2017). Secreted angiogenic factor that is almost exclusively expressed by endothelial cells and stimulates tumor angiogenesis in a manner similar to VEGF expressed by non-endothelial cell types (such as fibroblasts). miR-126 inhibits the proliferation, migration, and angiogenesis of RCC cells by downregulating EGFL7 expression and inhibiting the downstream ERK/STAT3 signaling pathway. lncRNA HOTAIR, as a competing endogenous RNA (ceRNA), may promote RCC development (Guo et al., 2021). RNA interference directed against VEGF or EGFL7 has been shown to attenuate angiogenesis and tumor growth in RCC (Gu et al., 2015; Xu et al., 2014).

Beyond VEGF, there are also angiogenesis-related genes such as tyrosine-protein kinase receptor-2 (TIE2), angiopoietin 2 (ANG2) and caveolin-1 (CAV1). The oncogenic protein Myoferlin (MYOF) upregulates these genes and promotes RCC migration (Cox et al., 2020). Estrogen receptor β (ERβ) promotes RCC progression by altering angiopoietin (ANGPT-2)/Tie-2 signal-mediated angiogenesis. Therefore, using the anti-estrogen drug ICI182,780 to inhibit ERβ/ANGPT-2/Tie-2 signaling can increase the sensitivity of sunitinib and better inhibit the progression of ccRCC (Gu et al., 2020).

In addition to the regulation of canonical angiogenic genes such as VEGF and TIE2, multiple alternative mechanisms have been identified that govern RCC neovascularization. For example, Branched-chain keto-acid dehydrogenase kinase (BCKDK), the rate-limiting enzyme of branched-chain amino acid (BCAA), the rate-limiting enzyme of branched-chain amino-acid catabolism, has been shown to enhance angiogenesis by up-regulating exosomal miR-125a-5p and concomitantly down-regulating its downstream targets (Yang et al., 2023). Conversely, E74-like transcription factor 5 (ELF5), which belongs to the E26 transformation-specific (ETS) family, stabilizes USP3 through transcriptional activation, significantly weakens angiogenic ability, and acts as a tumor suppressor gene. It is worth mentioning that ELF5 is downregulated in RCC (Ding et al., 2024).

VHL can act as an E3 ubiquitin ligase to ubiquitinate the α subunit of HIF in an oxygen-dependent manner; under normoxic conditions, this binding can degrade the α subunit of HIF. However, under hypoxic conditions, HIF is not degraded, and the HIF signaling cascade promotes angiogenesis. Recently, the role of the VHL-HIF pathway in RCC has been emphasized, and HIF-2α inhibitors have also become a treatment option for patients with metastatic RCC (Roviello et al., 2024). HIF-1α itself behaves as an oncogene that accelerates tumor growth and dissemination by sustaining hypoxia-responsive signaling and up-regulating HIF-2α target genes, including cyclin D1, c-Myc, VEGF-A, EGFR, TGF-α, and GLUT-1 (Mazumder et al., 2023). Gramicidin A is a prototype channel-forming ion carrier that can also inhibit angiogenesis by inhibiting (David et al., 2014).

TKI-resistant RCC cells are highly susceptible to metastasis. They reduce miR-549a through the VEGFR2-ERK-XPO5 pathway, which in turn promotes the expression of HIF-1α, resulting in increased VEGF secretion and further activation of VEGFR2 to form a feedback effect (Xuan et al., 2021). p53 downregulates HIF-1α in RCC, while p53 overexpression reduces VEGF production. The increase of transglutaminase 2 (TGase 2) promotes angiogenesis by inducing p53 degradation, leading to HIF-1α activation. The interaction between HIF-1α and p53 with cofactor p300 is required for stable transcriptional activation. TGase 2-mediated p53 depletion promotes angiogenesis by increasing HIF-1α-p300 binding in RCC (Lee et al., 2020). Polypyrimidine tract-binding protein 1 (PTBP1) is a heterogeneous ribonucleoprotein that can inhibit cell migration, invasion, and angiogenesis in vitro and lung metastasis of RCC in vivo through (Shan et al., 2018). ELR510444 is a promising new hypoxia-inducible factor inhibitor that reduces the expression of both HIF-1α and HIF-2α, inhibits angiogenesis, and has vascular destructive properties (Carew et al., 2012).

lncRNA-ECVSR can enhance ERβ expression and upregulate HIF-2α to exert its effects. Sunitinib is a first-line drug for the treatment of RCC, and it mainly works by inhibiting tumor angiogenesis. It is worth noting that the lncRNA-ECVSR/ERβ/Hif2-α signaling pathway increased by sunitinib leads to an increase in the cancer stem cell (CSC) phenotype, thereby promoting the formation of vascular mimicry (VM) (He et al., 2022). Migration and invasion inhibitory protein (MIIP) contains 388 amino acids and is considered to be a novel tumor suppressor that can inhibit RCC proliferation and angiogenesis by negatively regulating (Yan et al., 2021).

In summary, angiogenesis in RCC is a complex process involving the interaction of multiple molecules and signaling pathways (Table 2). A deeper understanding of these regulatory mechanisms will help provide new targets and strategies for the treatment of RCC.

Upon arrival in the stomach, RCC cells are accompanied by ECM remodeling that both shapes the local microenvironment and fosters tumor adaptation and expansion. Tumor cells can also destroy the ECM by secreting MMPs (Table 3), thereby promoting the infiltration and migration of tumor cells (Mao et al., 2021). MMPs, a family of 24 zinc-dependent endopeptidases whose proteolytic activity is counterbalanced by four tissue inhibitors of metalloproteinases (TIMP-1, TIMP-2, TIMP-3, and TIMP-4). They regulate the proteolytic activity of MMPs by blocking the active sites of MMPs. Both MMPs and TIMPs play an important role in ECM regulation and remodeling. Among them, MMP-9 is the most important and is highly expressed in metastatic RCC (Hashmi et al., 2021).

Increased expression of MMP-9 promotes RCC cell invasion and metastasis. G protein-coupled estrogen receptor (GPER), a non-genomic signalling mediator, has been shown to upregulate MMP-9 through downstream MAPK and PI3K/AKT. G-1 promotes RCC cell metastasis by activating GPER to enhance PI3K/AKT/MMP-9 signaling pathway (Guan et al., 2018). MUC15, a cell membrane-associated mucin in a variety of epithelial cells, has been demonstrated to inhibit EGFR and PI3K-AKT signaling pathways in vivo and in vitro, and downregulate MMP2 and MMP9 expression to inhibit invasive behaviors (Yue et al., 2020). Thymoquinone (TQ), a natural compound derived from Nigella sativa seeds, can also downregulate MMP-9 by inhibiting the activation of PI3K/AKT and p38, inhibiting the migration of RCC cells (Park et al., 2021). In addition to affecting tumor cell migration by directly acting on MMP-9, indirect mechanisms have been identified. NF-κB, a regulator of MMP-9 and a key transcription factor in cancer cells, has been implicated in tumor progression by regulating angiogenesis and inhibiting apoptosis, promoting tumor invasion and metastasis. For example, free fatty acid receptor-4 (FFA4), as a G protein-coupled receptor endogenously activated by medium- and long-chain free fatty acids, promotes the migration and invasion of ACHN cells through the PI3K/AKT/NF-κB signaling pathway (Karmokar and Moniri, 2023). Tumor necrosis factor-α (TNF-α) serves as an additional upstream regulator; the RGD (Arg-Gly-Asp)-toxin peptide has been reported to inhibit TNF-α-induced MMP-9 secretion, resulting in suppressed tumor-cell proliferation (Liu et al., 2021). Furthermore, Tropomyosin-1 (TPM1) has been found to upregulate E-cadherin expression while downregulating MMP-9 and VEGF expression, collectively inhibiting tumor metastasis in RCC (Wang et al., 2019). Distinct microRNA networks also converge on MMP-mediated pathways. Whereas miR-106b, miR-134 and miR-514a-3p suppress RCC invasion and migration through MAPK-related signaling, has-miR-15b-5p, has-miR-99b-5p, and has-miR-181a-5p have been demonstrated to interact directly with MMP-2 and MMP-9, thereby inhibiting RCC invasion, migration, and proliferation (Király et al., 2024).

Naturally occurring TIMPs regulate MMP-mediated proteolysis. Four inhibitors, TIMP-1, TIMP-2, TIMP-3, and TIMP-4, are primarily involved in inhibiting MMPs, thereby preventing cell migration, invasion, and angiogenesis (Koochekpour et al., 1999). Consistent with this function, elevated TIMP expression is associated with reduced tumor growth. Plant-derived flavonoids—ubiquitous polyphenols—have been demonstrated to upregulate TIMPs and, simultaneously, to modulate PI3K-AKT, MAPK, NF-κB, signal transducer and activator of transcription 3 (STAT3), and focal adhesion kinase (FAK) signaling cascades that govern cell motility and metastasis (Cayetano-Salazar et al., 2022). FK-506-binding protein 51 (FKBP51) is a high molecular weight protein encoded by the FKBP5 gene, and FKBP51 can promote the autophagic degradation of RCC (Mao et al., 2021).

In addition to MMPs, TIMPs and other important components of the ECM such as IL-2, IFN-α, Decorin (DCN), SMOC2 (SPARC-Related Modular Calcium-Binding), which affect the migration of RCC cells through the cytoplasmic matrix. DCN, a small leucine-rich proteoglycan synthesized by fibroblasts, has been shown to upregulate p21 and E-cadherin, thereby restraining proliferation and migration (Xu et al., 2016). SMOC2 is a matrix cell protein that enhances EMT through the integrin pathway and promotes the growth of RCC cells in the tumor (Feng et al., 2022). High concentrations of IL-2 and IFN-α produced locally at the RCC tumor site directly change tumor characteristics associated with the invasion and metastasis phenotype of RCC (Hathorn et al., 1994). Integrin-linked kinase (ILK) is a serine/threonine kinase that downregulates the EMT-related transcription factors Snail and Zeb1 to regulate the expression of vimentin and E-cadherin to inhibit invasion and metastasis (Han et al., 2015).

In conclusion, RCC cells provide favorable conditions for their adaptation, growth, infiltration and migration in the tumor microenvironment by changing the composition and regulatory mechanism of ECM. This ECM-centred reprogramming constitutes a pivotal mechanism underlying gastric metastasis of RCC.

Extracellular vesicles, notably exosomes, have emerged as pivotal mediators of inter-cellular communication within the tumor microenvironment, ferrying DNA, RNA, proteins, and metabolites that dictate recipient-cell fate (Table 4) (Flora et al., 2023; Rui et al., 2023). Communication between cancer cells that express tumor suppressor genes and those that do not is crucial for the distant metastasis of RCC. For example, in a new metastasis model, VHL not only promotes RCC metastasis by regulating VEGF through HIF to affect angiogenesis, but also exosomes produced by VHL (−) RCC cells can induce EMT, migration, invasion and metastasis in VHL (+) RCC cells (Schoenfeld et al., 2023). Macrophage-derived exosome miR-342-3p inhibits NEDD4L, which subsequently inhibits CEP55 ubiquitination and degradation by activating the PI3K/AKT/mTOR signaling pathway, upregulating CEP55 protein expression, and promoting the growth and metastasis of RCC cells (Feng et al., 2021). Similarly, Macrophage-derived exosomes delivering miR-193a-5p promote the progression of RCC by mimicking TIMP2-dependent angiogenesis (Liu et al., 2022).

Tumor cells release extracellular vesicles (EVs) containing several types of molecules, including small noncoding RNAs (sncRNAs) such as miRNAs, piRNAs, or tRNAs (Bogusławska et al., 2025). Within this network, RAB27A—a small GTPase that orchestrates exosome secretion—has been shown to drive the export of miR-127-3p, thereby accelerating RCC metastasis (Song et al., 2024). RAB27B has an oncogenic role in RCC cell lines, but its specific role in exocytosis has not yet been determined (Tsuruda et al., 2020). In hypoxic niches, tumour-associated macrophage-derived exosomes deliver miR-155-5p, which activates the IGF1R/PI3K/AKT cascade and promotes RCC progression (Gu et al., 2021). PiRNAs are sncRNAs with a length of 21–35 nucleotides that regulate gene expression through multiple mechanisms. Mesenchymal stromal cells (MSCs), integral components of the tumor microenvironment, have been reported to respond to the RCC-secreted piRNA piR_004153; this sncRNA alters MSC expression of FGF2, SLC7A5 and WISP1 and enhances MSC migration and survival, providing the first demonstration that cancer-derived piRNAs can augment stromal cell motility (Bogusławska et al., 2025).

Apolipoprotein C1 (APOC1) is the smallest apolipoprotein (6.6 kDa) and is a component of triglyceride-rich lipoproteins and high-density lipoproteins. Macrophages overexpressing APOC1 in RCC promote the metastasis of RCC cells by secreting CCL5 (Ren et al., 2022). APOC1 is upregulated as an oncogene in RCC and promotes the proliferation, migration and invasion of RCC by activating Wnt3a signaling pathway (Jiang et al., 2021). Furthermore, ApoC1 in exosomes is transferred from ccRCC cells to vascular endothelial cells and promotes the metastasis of ccRCC cells by activating STAT3 (Li et al., 2020).

This review introduces the mechanism underlying RCC metastasis to the stomach, integrating PI3K/AKT-driven survival, Ras/MAPK-mediated motility, and Wnt-orchestrated metabolic plasticity with concurrent angiogenic reprogramming, extracellular-matrix remodeling, and exosome-mediated signaling. Although the precise mechanism remains unresolved, the outlined pathway network provides a rational framework for targeted intervention; further investigation is required to clarify these processes.