Gastric Epithelial Cell Plasticity and Molecular Mechanisms of Metaplastic Transformations in the Stomach

Abstract

This research delves into the complex relationship between progenitor cells and the differentiated cell types that make up the stomach. It highlights the need for further investigation into the mechanisms governing stomach development and how these mechanisms relate to the maintenance of the stomach in a healthy state. The transition from normal gastric mucosa to metaplasia involves significant alterations in the phenotype and function of gastric epithelial cells, including stem cells, mucous neck cells, chief cells, and parietal cells. The presented literature review provides an in-depth analysis of pyloric and pseudopyloric metaplasia, along with spasmolytic polypeptide-expressing metaplasia, focusing on their biological significance, underlying pathogenesis, diagnostic features, and prognostic implications. It explores the role of various gastric epithelial cell types in the pathogenesis of metaplasia, highlighting recent advances in cellular plasticity, molecular pathways, and the implications for gastric carcinogenesis.

1. Introduction

Despite some advances in cancer treatment, gastric carcinoma remains one of the leading causes of cancer-related deaths globally [1,2]. Gastric cancer [GC] is preceded by precancerous lesions, including dysplasia, intestinal metaplasia (IM), and atrophy [3]. The stomach, as one of the organs with the highest degree of self-renewal in the body, exhibits significant cell plasticity. Differentiated mature cells can be reprogrammed into another mature cell as a reserve stem cell to replace lost cells. This plasticity implies that gastric epithelial cells may undergo cycles of differentiation and dedifferentiation, thereby increasing the risk of accumulating cancer-prone mutations [4]. The dysregulation of cellular plasticity in gastric epithelial cells can lead to the development of precancerous lesions, which can progress to GC. Understanding the mechanisms of plasticity and its role in the development of precancerous lesions is crucial for developing targeted therapies and preventive strategies for GC. Although numerous studies have been conducted, the molecular pathways that drive precancerous alterations, particularly metaplastic changes, are still not well elucidated [5,6].

A comprehensive study of cellular development in the normal stomach and molecular mechanisms of cellular plasticity should help us better understand the origins of gastric precancerous lesions, such as metaplasia and dysplasia, one of the most common precursors of GC [7,8]. That is why the main goal of this review is to illuminate the molecular mechanisms underlying metaplasia. It explores the cellular origins of metaplasia, potential mechanisms for the transition between cell types, and the role of metaplasia in disease progression, particularly the development of dysplasia and GC.

2. Terminology and Definitions

The area of research discussed in this review is still developing, and the terminology used is not yet consistent. We aim to use terms as clearly and precisely as possible, recognizing that, in many cases, there is not enough data to definitively label certain cellular phenomena. We will use terminology that we believe is most suitable and aligns with how most researchers in the field currently use it.

2.1. Cell Plasticity

Cell plasticity refers to the ability of cells to modify their phenotypes in response to environmental signals without altering their genotypes [4,9]. Depending on the nature and direction of cell type transitions, cell plasticity can manifest in several forms: transdetermination, transdifferentiation, transcommitment dedifferentiation, reversion, paligenosis, and metaplasia. The majority of these processes involve cellular reprogramming, a mechanism by which cells alter their identity and function.

2.1.1. Transdetermination

Transdetermination classically refers to the conversion of one progenitor/stem cell population into another, thereby potentially forming a basis for a metaplastic tissue transformation. This phenomenon has been most thoroughly characterized in Drosophila, where repeated leg-to-wing transformations were observed following transplantation of the imaginal disk. Comparable transdetermination events can also be triggered directly within tissues by ectopic activation of the wingless gene, implying that while transdetermination is an intrinsic property of certain cells, it can be externally stimulated. In contrast, this process remains largely unexplored in vertebrate systems [9].

2.1.2. Transdifferentiation

Transdifferentiation (lineage reprogramming), in contrast to transdetermination, refers to the conversion of one differentiated cell type into another with or without the involvement of an intermediate progenitor-like stage, thereby affording another possible mechanism for tissue-level metaplasia [10]. It has been demonstrated in some cells, like the transformation of pancreatic cells into hepatocytes [11].

Current evidence suggests that most, if not all, cells undergoing this transformation transition through an intermediate phase, implying that the notion of “direct” transdifferentiation is more intricate and less linear than originally believed [12,13]. During transdifferentiation, cells were thought to pass through transitional states resembling those seen during development [14]. The progression of cells through intermediate states is not yet fully understood. Given the potential application of transdifferentiation in regenerative medicine and disease modeling, a better understanding of intermediate states is crucial to avoid uncontrolled conversion or proliferation, which poses a risk for patients. Overall, transdifferentiation is a complex process, and our understanding of transitional states at the single-cell level is still limited [14].

The most extensively studied form of transdifferentiation is the epithelial-to-mesenchymal transition (EMT), a process in which epithelial cells lose their apical-basal polarity and cell–cell adhesion properties and acquire mesenchymal features, including enhanced motility and invasiveness, elaborated by Kalluri and Weinberg [15]. Studies by Nieto and colleagues have further defined EMT as a key regulator not only of developmental processes but also of pathological conditions such as fibrosis and tumor progression [16].

The reverse process, mesenchymal-to-epithelial transition (MET), also plays a central role in tissue regeneration and cellular reprogramming, as highlighted in the work of Mani et al. [17]. Together, EMT and MET represent fundamental mechanisms of cellular plasticity, especially in the context of cancer progression, metastasis, and therapeutic resistance, as demonstrated in studies by Dongre and Weinberg [18]. These transitions exemplify how transdifferentiation is not a linear or direct process but often involves reversible and dynamic phenotypic states.

Several signaling pathways, including TGFβ, Wnt, Notch, Hedgehog, and Hippo, are involved in EMT. Modifications to the extracellular matrix (ECM) and cell interactions with it also influence EMT [19,20]. Criteria to define complete cell conversion to a new fate are unclear (i.e., molecular, functional, or both) [20,21].

2.1.3. Transcommitment

Transcommitment is molecular reprogramming of a stem or progenitor cell, while transdifferentiation is molecular reprogramming of a fully differentiated cell [22]. Progenitor cells undergoing transcommitment involve molecular reprogramming through altering the expression of various transcription factors. This process can lead to changes in cell fate and specialized differentiation pathways, potentially influencing conditions like Barrett’s metaplasia [23].

2.1.4. Dedifferentiation

Dedifferentiation means that a differentiated cell reverses its differentiated fate to acquire properties it had previously during development (i.e., it returns to a state resembling the progenitor, precursor, and/or stem cell state it passed through to become a differentiated cell). The strictest sense of the term “dedifferentiation” carries the connotation that a differentiated cell not only reverts to a precursor or progenitor state but that it also acquires the capacity for multipotency, in other words, to differentiate into cells of a different lineage [24].

Mechanistically, when a differentiated cell reverts to its parental lineage or less-differentiated cell to acquire a proliferative phenotype, while transdifferentiation suggests the direct conversion of a differentiated cell type to another differentiated cell type without entering a pluripotent state.

Transdetermination and transdifferentiation involve lateral changes between distinct, specialized cell types. In contrast, dedifferentiation refers to an upward shift in cellular hierarchy, where mature cells regress to a progenitor-like or less differentiated state [25]. A prominent example of this process is the induction of pluripotent stem cells (iPSCs) from fully differentiated cells via epigenetic reprogramming [26]. Dedifferentiation can also be triggered by pathological stimuli, such as tissue injury, where cells revert to a progenitor state, proliferate, and differentiate to repair the affected tissue [9]. The term is also used by pathologists to describe cancer phenotype unrelated to developmental reversion.

2.1.5. Reversion

Reversion largely appears to be synonymous with dedifferentiation, though reversion seems to be used more specifically to denote return specifically to the adult tissue stem cell state vs. return to progenitor or embryonic states. It is still unclear how reversion differs from dedifferentiation. Other terms describing cell plasticity include “reversion,” which has been used to denote when a mature or quiescent cell returns to a stem cell state [27]. It seems to be used predominantly by scientists studying adult differentiation and injury response in the intestine, though there is precedent for reversion as a general term meaning dedifferentiation [21].

2.1.6. Paligenosis

Paligenosis is a conserved, stepwise cellular program through which mature, differentiated cells re-enter the cell cycle and acquire progenitor-like characteristics in response to injury or stress. This process enables tissue regeneration by activating autophagy, suppressing differentiation genes, and reactivating proliferative programs [28].

Paligenosis was proposed to define the specific, evolutionarily conserved molecular machinery cells use to dedifferentiate or transdifferentiate. Paligenosis seems to be required for metaplasia, a precancerous lesion that occurs during chronic injury and inflammation [21].

2.1.7. Cellular Reprogramming

Cellular reprogramming refers to the process of converting specialized, differentiated cells back into a pluripotent state, creating iPCSs. This was first accomplished in 2006 by Takahashi and Yamanaka, who showed that introducing a specific set of transcription factors (Oct4, Sox2, Klf4, and c-Myc) could reprogram somatic cells like fibroblasts to resemble embryonic stem cells [26]. Their pioneering work transformed regenerative medicine and paved the way for advancements in disease modeling, drug development, and personalized treatment approaches.

During cellular reprogramming, which involves changing the identity of a cell, both transcriptional and epigenetic changes occur. Besides the changes in gene expression (transcription), cells also undergo alterations in their epigenetic landscape, including modifications to DNA methylation and histone modifications. These changes are crucial for the reprogramming process and the establishment of a new cellular identity [29].

Reprogramming involves changes in histone modifications to create a more permissive chromatin structure for the new transcriptional program. The interplay between DNA methylation and histone modifications is crucial. For example, histone modifications can influence DNA methylation by recruiting or repelling enzymes involved in DNA methylation.

2.1.8. Metaplasia

The term ‘metaplasia’ refers to a wide class of cell-type transformations, including transdifferentiation, transdetermination, and, more recently, cellular reprogramming. The classical definition of metaplasia is “the conversion, during postnatal life, of one differentiated cell type to that of another” [30,31,32].

This term is typically described at the tissue level rather than at the cellular level. It remains uncertain whether the process is reversible or permanent. In summary, metaplasia is a broader term for any cell type transformation, while transdifferentiation is a specific type of metaplasia that occurs without involving a stem cell stage (Figure 1).

3. Gastric Epithelium Development and Renewal

The stomach is made up of tissues derived from all three embryonic germ layers: the ectoderm forms the enteric nervous system, the mesoderm develops into the smooth muscle and mesenchymal components, and the endoderm gives rise to the epithelial lining [28].

Based on histology, ultrastructure, and the specific substances they produce, five distinct differentiated cell types can be identified in the adult corpus—the primary functional region of the stomach. These include the following: foveolar (pit) cells at the tops of gastric glands, which produce mucus and renew every three days; zymogenic (chief) cells at the gland base, which secrete digestive enzymes like pepsinogen and have a turnover of several months; parietal (oxyntic) cells, found throughout the gland shaft, which produce hydrochloric acid; endocrine cells, comprising less than 2% of the epithelium, which release hormones; and tuft cells, which are equally rare, have poorly understood roles, and are marked by chemosensory proteins and distinctive apical microtubules. In the antrum, along with pit, endocrine, and occasional parietal cells, the gland base also contains cells that produce protective mucins [33]. Depending on the species, antral units may contain some chief and parietal cells, but they are primarily composed of mucus-secreting pit (foveolar) cells on the surface and deeper glandular cells that show features of both mucous neck and chief cells. Scattered throughout the antrum are rare endocrine cells, each identified by the main hormone they release—for instance, G cells that produce gastrin [28]. Each of these cell types is generated by stem and progenitor cells located in the isthmus of discrete gland units [33].

The gastric stroma comprises a heterogeneous population of mesenchymal-derived cells, including fibroblasts, myofibroblasts, pericytes, and telocyte-like cells. These are typically characterized by spindle-shaped nuclei and elongated cytoplasmic projections. Fibroblasts, in particular, synthesize components of the ECM such as collagen, proteoglycans, fibronectin, and metalloproteinases. In the gastrointestinal tract, fibroblasts are recognized for their role in preserving epithelial homeostasis and reacting to environmental cues from mucosal infections or injuries, and support the proliferative niche by releasing growth factors [34].

Sigal et al. discovered that stromal cells at the base of the stomach’s antral glands produce R-spondin (Rspo). Through various mouse models and fluorescence in situ hybridization, they showed that fibroblasts near the gland base express both α-smooth muscle actin (αSMA, encoded by the ACTA2 gene) and RSPO3. RSPO3 plays a crucial role as a niche-specific regulator, helping to maintain and control the growth of epithelial stem cells located at the base of the mouse antral glands [35].

Pastula et al. showed that when myofibroblasts were co-cultured with gastric organoids, the organoids’ growth was supported. They proposed that fibroblasts secrete niche factors essential for the proliferation of epithelial stem cells [36].

Recent single-cell RNA sequencing (scRNA-seq) studies have identified four distinct fibroblast populations in the normal human stomach: one with high PDGFRA expression (PDGFRA^hi), associated with genes involved in ECM organization and cell movement (telocyte-like fibroblasts); another with elevated FBLN2 expression (FBLN2^hi), linked to inflammatory pathway genes (inflammatory fibroblasts); a third population showing high ACTA2 and low PDGFRB expression (ACTA2^hi, PDGFRB^lo), enriched in myofibroblast-related genes (myofibroblasts); and a fourth with low ACTA2 but high PDGFRB expression (ACTA2^lo, PDGFRB^hi), enriched in genes typical of pericytes. Among these, the myofibroblast population expressing ACTA2 is the predominant one in the normal gastric mucosa, even in cases of gastritis. All four fibroblast populations are found in stomach tissues across different conditions, including normal, inflamed, metaplastic, and dysplastic or cancerous areas [34].

Key Transcription Factors and Signaling Pathways in Stomach Development

The development and differentiation of gastric epithelial cells is a highly intricate and dynamic process that occurs through several stages and is regulated by multiple mechanisms. Precursors of gastric epithelial cells originate from the foregut portion of the endoderm, where multipotent progenitor cells are controlled by crucial transcription factors such as FoxA, Gata, Sox17, and Mixl1, which play vital roles in the proliferation and preservation of foregut progenitors [4,37]. These factors are essential for forming foregut progenitor cells, which ultimately give rise to the stomach. As regionalization progresses, various transcription factors become either broadly expressed across the foregut or localized within specific pregastric regions. Factors such as Foxa1/2/3, Gata4/6, Hnf1b, and Sox2 are broadly expressed and are believed to play key roles in gastric identity. FoxA transcription factors, in particular, help promote the expression of Pdx1 in the foregut. Pdx1 is restricted to the gastric antrum and is absent from the more proximal corpus. Gata4 is thought to contribute to the development of the glandular stomach, while SOX2 plays a more critical role in specifying the anterior foregut and stomach than previously recognized. The expression boundary between Sox2 and Cdx2 marks the future gastroesophageal junction, suggesting that Sox2 may define this anatomical border [28].

Cells expressing the Sox2 promoter exhibit stem cell capabilities, including self-renewal and the ability to differentiate into all epithelial lineages within the gastric corpus. Sox2+ adult stem and progenitor cells are important for tissue regeneration and survival of mice [38]. SOX2 protein is present at mid-to-low levels in various cell lineages throughout the corpus and has also been used as a marker of gastric differentiation in humans, as opposed to intestinal differentiation [39]. Sox2 and Oct4 drive the reprogramming of trophoblast stem cells to pluripotency [38].

Pdx1 expression serves as a developmental marker, distinguishing antral progenitors (SOX2 + GATA4 + PDX1+) from corpus progenitors (SOX2 + GATA4 + PDX1−). Loss of Pdx1 leads to abnormal development of antral progenitors, resulting in pyloric defects and the absence of gastrin-secreting endocrine cells. Hnf1b, which is expressed in the early endoderm, has also been implicated in gastric development [4,28].

Additional transcription factors contribute to the specialization of gastric cell types: Spdef is crucial for the maturation of deep mucous cells in the antrum; Foxq1 regulates the expression of MUC5AC, the primary mucin secreted by pit cells; and Xbp1 and Mist1 (also known as Bhlha15) are vital for the structural maturation of chief cells [4,28].

The chief cells, typically considered fully differentiated, can also function as reserve stem cells and are capable of giving rise to all cell lineages within the corpus. Muscle, Intestine and Stomach Expression 1 (Mist1) MIST1 protein and its mRNA are almost exclusively expressed in mature chief cells, while TROY (Tumor Necrosis Factor Receptor Superfamily Member 19 (Tnfrsf19 or Troy)) expression is limited to a small subset of chief and parietal cells [40]. Differentiated Troy⁺ chief cells act as reserve stem cells to generate all lineages of the stomach epithelium. Furthermore, a population of Mist1-expressing cells located in the isthmus may represent another potential source of stem cells in the corpus, suggesting that multiple regions may harbor regenerative capacity [40].

The chief cells in the gastric epithelium do not form directly from epithelial progenitors but develop through an intermediate stage involving mucous neck cells. These intermediary cells, enriched with mucin-containing vesicles, originate from glandular stem cells and pass through precursor phases termed “pre-neck” cells. Located in the neck region of the gland, mucous neck cells migrate downward, interspersing with parietal cells, before arriving at the gland base, where they rapidly differentiate into mature chief cells. Mucous neck cells are distinct from chief cells in several ways: they express specific markers like Gkn3, Tff2, and Muc6; possess a characteristic morphology with narrow extensions reaching the basement membrane and broad apical domains facing the lumen; and serve a specialized function related to mucin production [41].

Ascl1 and Ngn3 are both essential for the development of endocrine cells that produce gastrin, somatostatin, and glucagon. Ngn3 serves as a marker for endocrine progenitor cells but is not expressed in fully differentiated endocrine cells. Populations expressing Ascl1 or Ngn3 give rise to specific endocrine cell types based on the downstream transcriptional programs they activate. Evidence suggests that similar regulatory mechanisms operate in the stomach, as mice lacking certain transcription factors exhibit defects in particular endocrine lineages. For instance, Pdx1, Nkx6.3, Pax4/6, and Arx have all been linked to the differentiation of specific mature gastric endocrine cell types [41].

The mesenchymal compartment that surrounds the gastric glands has received comparatively little attention and remains poorly characterized. Despite its close physical association with the epithelial cells, the specific cellular composition, signaling interactions, and functional roles of this stromal environment are not yet well defined [42]. A number of transcription factors—Bapx1 (Nkx3-2), Nkx2-5, Gata3, Six2, Nr2f2 (also known as Chicken Ovalbumin Upstream Promoter-Transcription Factor II), and Sox9—are expressed in the posterior mesenchyme of the developing stomach and are essential for the formation and specification of the pyloric region [43,44,45]. Disruption or absence of these transcription factors leads to defective neuromuscular patterning of the pyloric sphincter, a defect that can manifest in humans as pyloric stenosis, a relatively common gastrointestinal condition in infants [46].

While BMP signaling, primarily originating from the mesenchyme, plays a crucial role in gastric epithelial development, Hedgehog signaling, secreted by the epithelium, influences mesenchymal differentiation. For example, in addition to their early roles in foregut patterning, Sonic Hedgehog (Shh) and Indian Hedgehog (Ihh) are produced by the gastric endoderm and act to support mesenchymal growth and development, a signaling dynamic that is preserved in the adult stomach [47,48]. On the other hand, mesenchymal signals also impact epithelial differentiation. One such example is fibroblast growth factor 10 (FGF10), which appears to signal through FGFR2B to promote epithelial proliferation and gland formation [49]. Although FGF10 is not strictly essential for adult epithelial maintenance, it has been shown to suppress the differentiation of parietal and chief cells, favoring instead the mucous neck cell lineage [50]. Additionally, Barx1, a mesenchyme-restricted transcription factor, is expressed in the prospective esophageal and gastric mesoderm and plays an essential role in early regional patterning of the stomach [28,51]. While the exact identity of the adult corpus gastric epithelial stem cell remains elusive, early studies have begun to characterize the molecular and cellular differentiation patterns of mature cell lineages derived from this stem cell. Notably, several studies support a model where mucous neck cells differentiate into chief cells through a transdifferentiation [28,52]. Although the evidence for this lineage relationship is mostly indirect, it is substantial and varied. For instance, cells co-expressing neck and chief cell markers have been observed during postnatal development [53]. Moreover, mature gastric units harbor transitional cells exhibiting both neck and chief cell features, confirmed through ultrastructural and gene expression analyses [28,54,55]. Upon stomach injury, metaplastic cells derived from the chief cell lineage also express markers of both cell types [55,56,57]. Lineage tracing with the Tff2 promoter further suggests that parietal cells and mucous neck/chief cells originate from a common progenitor pool [58]. Additionally, deletion of Mist1 or Xbp1, key regulators of chief cell maturation, increases the prevalence of cells with mixed neck–chief characteristics [56,57].

Surface mucous (pit/foveolar) cells arise rapidly from progenitors located in the isthmus, although the exact nature of these progenitors is not fully established—they may be committed pit-specific progenitors, multipotent stem cells, or both. Interestingly, decreased proliferation in the isthmal progenitor zone preferentially affects pit cells over deeper glandular cells, suggesting that proliferation here mainly supports surface mucous cell replenishment under normal conditions [52]. The transcription factor FOXQ1 is important for pit cell differentiation because it regulates the expression of MUC5AC, a principal mucous granule protein, though it is not required for lineage specification itself [25].

Multiple signaling pathways regulate stomach homeostasis and cell behavior. Notch signaling is active in the corpus isthmus, promoting proliferation in this region. Ectopic activation of Notch in parietal cell lineages blocks differentiation and maintains progenitor-like features [59]. Inhibition of BMP signaling leads to increased cell proliferation in adult stomach glands, with a reduction in parietal cells and accumulation of transitional cells expressing both neck and chief cell markers at the gland base, highlighting BMP’s role in regulating progenitor proliferation and cell maturation in the corpus [28,60,61].

In the antrum, Notch signaling plays a key role in regulating the activity of Lgr5-positive stem cells. When Notch signaling is inhibited, differentiation toward mucous and endocrine cells is promoted, whereas activation of this pathway suppresses differentiation [53]. BMP signaling via the receptor BMPR1A controls both the proliferation and differentiation of antral mucous cells. Loss of BMPR1A leads to excessive proliferation of these mucous cells and a failure to express the mucin MUC5AC. Mutations in BMP family members are linked to juvenile polyposis syndrome, a condition characterized by polyps throughout the gastrointestinal tract, including the antrum [53]. Since LGR5 functions as a co-receptor for canonical WNT signaling and is expressed in the antrum, WNT signaling likely contributes to maintaining antral homeostasis [62]. Lgr5 expression is restricted to a specific antral cell population that can efficiently give rise to all cell lineages within the antrum and cardia, but not the corpus [62]. Similarly, lineage tracing using Cck2r marks +4 antral cells (a nomenclature borrowed from intestinal stem cell studies, referring to cells located four positions above the base) that also possess stem cell potential and can generate Lgr5-positive antral cells [63]. Additionally, rare cells in antral glands labeled by Villin- and Sox2-promoter-driven lineage tracing demonstrate stem cell-like properties. However, the relationships among these various stem cell populations remain unclear. Notably, Sox2 is expressed broadly in many cells and therefore may not mark a distinct stem cell population [28,38] (

Table 1. Stomach development key transcription factors and signaling pathways.

Category	Components/Cell Types/Genes	Key Functions and Notes	References
Early Gastric Development	Foregut Endoderm, Progenitor Cells	Originate from endoderm; regulated by FoxA, Gata, Sox17, Mixl1	[28,64]
	Transcription Factors: FoxA, Gata4/6, Sox2, Pdx1, Hnf1b	Guide early differentiation; FoxA promotes Pdx1; Pdx1 marks antral progenitors	[4,28]
	Regionalization	Corpus: SOX2+GATA4+PDX1−; Antrum: SOX2+GATA4+PDX1+	[4,28]
Cell Lineage Differentiation	Mucous Neck Cells → Chief Cells	Transitional lineage; neck cells express Tff2, Muc6, Gkn3	[28,33]
	Chief Cells	MIST1+, Xbp1+; arise from mucous neck cells; can act as reserve stem cells	[28,33,61]
	Pit/Foveolar Cells	Arise from isthmal progenitors; FOXQ1 necessary for MUC5AC production	[28,33,61]
	Parietal Cells	Derive from shared progenitors; regulated by Notch and BMP signals	[28,33,61]
	Endocrine Cells	Require Ascl1, Ngn3, Pdx1, Nkx6.3, Pax4/6, Arx for lineage specification	[28,33,61]
Stem Cells and Progenitors	Sox2+ Cells (Corpus)	Self-renewing; multipotent; most stem-like ones at base, not isthmus	[28,33,61]
	Mist1+, TROY+ Cells	Chief cell lineage with stem cell capacity	[28,33,61]
	Lgr5+ (Antrum), Cck2r+ (+4 cells)	Antral-specific stem cells; generate all antral lineages	[28,33,61]
	Villin+, Sox2+ (Antrum)	Rare, stem-like populations; lineage tracing shows multipotency	[28,33,61]
Key Transcription Factors	Spdef	Maturation of antral deep mucous cells	[28,33,61]
	Foxq1	Pit cell MUC5AC expression	[28,33,61]
	Xbp1, Mist1 (Bhlha15)	Chief cell maturation	[28,33,61]
	Barx1	Gastric mesoderm specification (esophagus/stomach boundary)	[28,33,61]
Signaling Pathways	Notch	Promotes progenitor maintenance; inhibits differentiation (corpus and antrum)	[28,33,59,61]
	BMP	Restricts proliferation, promotes parietal/chief cell fate	[28,33,61]
	Hedgehog (Shh/Ihh)	Epithelial → mesenchyme signaling; supports mesenchymal differentiation	[28,33,61]
	FGF10 → FGFR2B	Mesenchymal → epithelial signal; inhibits chief/parietal, promotes mucous cell fate	[28,33,61]
	WNT/LGR5	Important in antral stem cell regulation	[28,33,61]
Mesenchymal Transcription Factors	Bapx1, Nkx2-5, Gata3, Six2, Nr2f2, Sox9	Pyloric specification; absence → pyloric stenosis	[28,33,61]
Metaplasia and Injury Response	Neck–chief transitional cells, re-expression of progenitor markers	Transitional phenotypes in injury/metaplasia; evidence for plasticity	[28,33,61]

).

Numerous aspects of stomach specification remain poorly understood or have yet to be thoroughly explored. Although it is clear that interactions between the gastric endoderm and mesoderm are fundamental for gastric development, the exact mechanisms that guide naive endodermal progenitors toward a gastric progenitor fate are still largely unknown. Additional research is needed to determine which cell populations in the adult stomach—particularly within the corpus—possess stem cell properties and to elucidate their specific roles in maintaining tissue homeostasis under normal conditions as well as their responses to injury or disease [28].

4. Metaplastic Transformations in the Gastric Mucosa

Precancerous lesions of GC represent a crucial pathological stage in the development of intestinal GC. Early detection and diagnosis are key to reducing the incidence of GC [65]. Although dysplasia in the gastric mucosa has been extensively investigated [66,67], metaplastic alterations have received comparatively less attention in research [68,69]. This gap highlights the need for more focused studies on metaplasia, as understanding these early cellular changes is crucial for unraveling the progression toward malignancy and developing effective prevention strategies [70].

Metaplastic changes in the gastric mucosa refer to the transformation of one differentiated gastric epithelial cell type into another, often in response to chronic injury or inflammation. Currently, two primary types of metaplasia in the stomach are distinguished based on their cellular characteristics and clinical implications:

1.

Spasmolytic Polypeptide-Expressing Metaplasia (SPEM)

SPEM is characterized by mucous cells expressing spasmolytic polypeptide (TFF2), typically emerging in the gastric corpus following parietal cell loss. It often develops as a reparative response to chronic injury such as Helicobacter pylori (H. pylori) infection or autoimmune gastritis. SPEM cells exhibit features similar to mucous neck cells and antral gland cells and are considered potential precursors to dysplasia and GC under persistent injury conditions [4,5,71].

2.

Intestinal Metaplasia

IM involves the replacement of the gastric epithelium with intestinal-type cells, including goblet cells and absorptive enterocytes. IM is further classified into complete and incomplete subtypes based on mucin expression and cellular differentiation. IM is a well-established premalignant lesion closely associated with chronic H. pylori infection and a significant risk factor for progression to gastric adenocarcinoma [72].

Metaplastic lesions in the stomach are known to be unequivocally associated with intestinal-type GC; however, the reversibility of these lesions remains controversial [72,73].

4.1. Spasmolytic Polypeptide-Expressing Metaplasia (SPEM, Pseudopyloric Metaplasia, Antralization)

4.1.1. SPEM Terminology

Pyloric metaplasia and pseudopyloric metaplasia (PPM) share similar histological appearances under H&E staining, but differ in immunohistochemical profiles—pyloric metaplasia is MUC6-positive and PGI-negative, while PPM is PGI-positive. Despite these differences, many studies group them together under general terms like “pyloric or PPM,” though accurate terminology is increasingly emphasized, especially distinguishing these from SPEM (spasmolytic polypeptide-expressing metaplasia). Wada et al. argue that SPEM, originally described in animal models, should not be used for human tissue due to limited TFF2 expression, and they demonstrate that gastrin-positive cells are found in pyloric metaplasia but not in PPM. SPEM lesions resemble the antral gland architecture but lack key features like PDX1 expression and gastrin-secreting G cells, aligning them more closely with embryonic stomach patterns [74,75,76]. Interestingly, the architecture of SPEM units also parallels that of the embryonic or neonatal stomach, which similarly lacks distinct parietal, neck, and chief cell populations [38,77]. Pseudopyloric or pyloric metaplasia, often seen along the greater curvature into the oxyntic region, is sometimes viewed as “antralization”—a process linked to aging, chronic gastritis progression, and the potential development of intestinal metaplasia and neoplasia through field cancerization [78]. SPEM lesions closely resemble antral glands at the cellular and molecular level, often displaying a superficial layer of TFF1- and MUC5AC-expressing foveolar cells overlying the metaplastic cells [79,80].

SPEM closely mimics the deep glands of the antrum, often displaying a luminal layer of TFF1- and MUC5AC-expressing foveolar cells above the metaplastic cells, replicating the architecture and cellular composition of antral glands [81]. A distinguishing molecular marker for SPEM, absent in the antral mucosa, is the protease inhibitor HE4 (WFDC2) [82,83]. Consequently, SPEM may be best understood as the molecular representation of what has historically been termed pseudopyloric or pyloric metaplasia [84]. In the context of this review, PPM and SPEM may be regarded as interchangeable, as Goldenring notes that “SPEM likely reflects the molecular definition of what has traditionally been described histologically as pseudopyloric or pyloric metaplasia.” [84,85] (

Table 2. Distinguishing features of pyloric, pseudopyloric, and SPEM in the stomach.

Feature	Pyloric Metaplasia	Pseudopyloric Metaplasia	SPEM	References
Histology (H&E)	Resembles native pyloric glands	Resembles pyloric glands	Resembles pyloric glands with altered gland base	[41,86]
MUC6	Positive	Positive	Strongly Positive	[41,52]
Pepsinogen I (PGI)	Negative	Positive	Variable; often downregulated	[41,82,86]
TFF2	Low/absent	Low	Strongly Positive	[41,82,86,87]
GKN3	Absent	Absent or low	Often present in mice	[52,82,88,89]
PDX1	Positive	Variable	Usually negative	[90,91,92]
Location	Antrum or transitional zones	Corpus, post-injury	Corpus/fundus, post-injury	[41,86]
Associated with injury	Not typically	Yes	Strongly associated with parietal cell loss	[41,86]
Lineage origin	Normal antral mucous cells	Reprogrammed chief cells	Reprogrammed chief (zymogenic) cells	[41,82,87]
Functional role	Normal mucous production	Adaptive repair response	Repair and pre-neoplastic potential	[41,89]
Clinical relevance	Benign	Precedes IM	Considered a preneoplastic lesion	[41,87,89]

).

4.1.2. Global Prevalence of SPEM

Since SPEM was first identified in 1999 [84], its morphological features and cellular origins have been clarified by various studies using mouse models and human tissues [52,86]. Though large-scale epidemiological data are limited, SPEM is increasingly recognized as a frequent metaplastic lesion in individuals with chronic gastric injury, particularly H. pylori-associated atrophic gastritis. For example, studies in Icelandic patients have identified SPEM in up to 82% of cases with early gastric neoplasia, compared to 37% of patients without neoplasia [73,93]. SPEM has also been observed in 61–88% of gastric remnant cancer biopsies, underscoring its frequent association with neoplastic progression [73].

4.1.3. SPEM: Mechanisms of Development and Biomarkers

Molecular and cellular mechanistic studies have shown that chronic atrophic gastritis (CAG) is not characterized simply by a chronic inflammatory infiltrate (gastritis) and the loss of acid-secreting parietal cells (oxyntic atrophy), but also by changes in differentiation of the chief cells (metaplasia) [28,82]. A thorough understanding of the processes that control the specification of cells within the gastric epithelium during development and adult homeostasis could be crucial to deciphering the disease etiology, particularly the metaplastic changes that arise after H. pylori infection [28]. Experimental lineage tracing and injury models have shown that SPEM can arise from both proliferating isthmal cells and transdifferentiated chief cells, though their relative contributions may vary depending on the type and severity of injury [94]. For example, following acute parietal cell loss (e.g., via DMP-777 or L-635 treatment), chief cells have been shown to directly convert into TFF2+ metaplastic cells without passing through a proliferative intermediate [52]. In contrast, chronic injury models, such as Helicobacter infection, induce both isthmal stem cell expansion and paligenotic remodeling of chief cells, leading to a more heterogeneous metaplastic response [95,96].

In humans, the development of SPEM is commonly triggered by chronic inflammation and oxyntic atrophy—the loss of parietal cells—often associated with H. pylori infection. To replicate this in experimental models, Petersen et al. and Goldenring et al. have employed drug-induced mouse models to produce acute parietal cell loss and mimic human SPEM [89,96,97]. Additionally, the Epstein–Barr virus has been implicated as another potential inducer of SPEM [98]. In parallel, emerging evidence has introduced the concept of “paligenosis”, a conserved cellular program through which terminally differentiated, post-mitotic cells—particularly chief (zymogenic) cells—can re-enter the cell cycle and initiate a regenerative transcriptional cascade in response to tissue injury. This phenomenon facilitates cellular reprogramming and metaplastic remodeling of the gastric epithelium [99,100]. The paligenotic pathway appears to be tightly regulated and is implicated in the formation of SPEM following parietal cell loss, positioning it as a central mechanism in gastric epithelial plasticity and potential neoplastic progression. Paligenosis, a process where mature cells revert to a progenitor-like state, involves three sequential stages: autodegradation, progenitor gene expression, and cell cycle re-entry. These stages are marked by distinct molecular and cellular changes, allowing for a controlled transition back to a proliferative state. These three stages are not isolated events but rather a coordinated program governed by specific molecular factors. The entire process can be blocked or reversed at any stage, making it a highly regulated and conserved cellular mechanism [101].

Severe injury to the corpus mucosa triggers the secretion of IL-33 by surface mucous cells and IL-25 by tuft cells, which together promote IL-13 production from group 2 innate lymphoid cells (ILC2s) [102,103]. This IL-13 is essential for initiating the reprogramming of chief cells into SPEM cells and for the subsequent formation of pyloric metaplasia [104,105].

Loss or injury of parietal cells leads to the downscaling of chief cells’ large secretory granules, which normally contain digestive enzymes such as pepsinogen C and carboxypeptidase B [52]. In response, these chief cells begin to re-express markers characteristic of mucous neck cells—their normal precursors in the adult stomach—including TFF2 (spasmolytic polypeptide), MUC6, the GS-II lectin-binding epitope, and, in mice, Gastrokine 3 [86,94,106]. Thus, parietal cell injury results in a broad reorganization of the gastric unit, marked by parietal cell atrophy, expansion of a metaplastic population that co-expresses both neck and chief cell markers, and the reprogramming of chief cells toward this metaplastic lineage [101,107]. Recent findings have demonstrated that CD44v9, a splice variant of the CD44 cell surface glycoprotein, is also strongly expressed in these cells [106,108,109]. However, using a single marker to diagnose SPEM can be problematic, as many of these proteins are not unique to SPEM. For instance, TFF2 and MUC6/GSII are also expressed by normal mucous neck cells in the gastric corpus, and CD44v9 may be present in both proliferating epithelial cells and IM lesions [110].

Recent research has identified Aquaporin 5 (AQP5) as a specific and early marker of SPEM, a precursor lesion in gastric carcinogenesis. AQP5, a membrane-bound water channel protein, was shown to be upregulated in SPEM cells during early stages of metaplasia in both mouse models and human tissues [73].

Importantly, in human gastric tissues, AQP5-positive cells were observed at the base of glands that also exhibited features of incomplete intestinal metaplasia (IIM)—a subtype known to carry a higher risk for GC. These cells were absent in glands with complete intestinal metaplasia (CIM), suggesting that AQP5 marks a transitional metaplastic state between SPEM and IIM [73,111]. Furthermore, the presence of AQP5 in basal gland cells, alongside Trop2 expression in the upper glandular compartment, reflects a state of lineage ambiguity within IIM, which may predispose tissues to dysplastic progression [73]. AQP5 not only serves as a diagnostic marker for early metaplasia but may also have oncogenic roles. Studies show that AQP5 overexpression promotes proliferation, migration, and invasion in gastric cancer cells, while its inhibition suppresses these behaviors [112,113]. Moreover, H. pylori infection has been shown to upregulate AQP5, facilitating EMT via the MEK/ERK signaling pathway, which links chronic inflammation to gastric tumor progression [113]. The combination of immunostaining for AQP5, Trop2, CD44v9, and TFF3 can therefore rapidly define the range of lineages present in the metaplastic stomach of human patients [73].

4.1.4. SPEM Classification

SPEM can be classified based on histological features, molecular expression patterns, and transitional potential, as detailed below:

• Morphological Classification
  • Mature SPEM
    Characterized by columnar mucous cells resembling deep antral or Brunner’s gland-like cells, located primarily at the base of oxyntic glands. These cells contain pale cytoplasm and basally located nuclei [114].
  • Proliferative/Active SPEM
    Displays increased proliferative index (Ki-67 positivity) and is often associated with inflammatory infiltrates and cytokine signaling, such as IL-13 [115]. It is also associated with pS6 expression [114,116].

2.

Molecular/Immunohistochemical Classification

SPEM is primarily defined by the expression of a core set of markers:

• TFF2 (Trefoil Factor Family 2)—A defining marker of SPEM, also known as spasmolytic polypeptide [82].
• MUC6—A mucin typically associated with deep gastric glands and mucous neck cells [82].
• GSII (Griffonia simplicifolia lectin II)—Binds to N-acetyl-D-glucosamine and is used as a glycoprotein marker of SPEM [117].

Additional markers involved in SPEM classification and progression include:

• CD44v9—Associated with cellular resilience, stemness, and GC risk [73].
• Aquaporin 5 (AQP5)—A recently identified early and lineage-specific marker of SPEM that helps identify transitional states toward IM [73].
• TROP2—Enriched in SPEM glands with IIM, marking a key progression interface [73].
• Clusterin—Identified in transitioning glands, potentially linked to chronic injury and metaplastic evolution [117].

3.

Transitional vs. Non-Transitional SPEM

• Transitional SPEM
  Exhibits co-expression of SPEM and IM markers, such as TFF2, MUC6, and TROP2, often seen at the base of glands where metaplasia is actively evolving [73]. Proposed to represent a critical intermediate step in the progression to IIM and gastric dysplasia [118,119,120].
• Non-Transitional SPEM
  Displays a stable secretory phenotype, retaining expression of MUC6 and TFF2, without markers of intestinal differentiation. More common in early or reversible injury settings [82].

4.

Induced vs. Spontaneous SPEM

• Induced SPEM
  Studied in experimental models involving parietal cell loss through agents such as DMP-777, L-635, or chronic H. felis infection. Provides insight into the cellular origin and molecular dynamics of metaplasia [118].
• Spontaneous SPEM
  Arises naturally in the setting of chronic gastritis, such as from H. pylori infection or autoimmune gastritis, and can progress to IM and GC [118,121,122].

Recent studies highlight stromal niche factors, particularly R-spondin 3 (RSPO3), as pivotal in regulating epithelial signaling and metaplastic progression [35,120]. RSPO3 is secreted by αSMA-positive stromal myofibroblasts located at the base of gastric glands and enhances Wnt–β-catenin signaling through binding to LGR receptors and inhibiting ubiquitin ligases RNF43/ZNRF3 [35]. In the antrum, this niche supports AXIN2⁺LGR5⁻ isthmal stem cells, promoting epithelial proliferation while modulating LGR5⁺ cell expansion [123,124,125].

RSPO3 in Response to Injury and Metaplastic Change: Upon injury, such as loss of chief and parietal cells, RSPO3 expression rises in stromal cells, activating YAP signaling in corpus glands and aiding regeneration of secretory lineages [123,126]. In chronic H. pylori infection, sustained RSPO3 expression maintains hyperproliferation and creates a niche conducive to metaplasia [113]. Beyond regeneration, RSPO3 promotes differentiation of LGR5⁺ cells into antimicrobial secretory cells expressing intelectin-1, contributing to bacterial clearance [126]. Loss of RSPO3 correlates with increased H. pylori colonization in gastric glands [126]. Crosstalk with Wnt, inflammation, and Oncogenesis RSPO3 amplifies Wnt–β-catenin signaling, expanding stem cells in metaplastic and neoplastic contexts. Chronic H. pylori infection enhances RSPO3 and Wnt pathway activity, leading to glandular hyperplasia and preneoplastic transformation [127,128]. These pathways cooperate with inflammatory signals and bacterial virulence factors, such as CagA-mediated β-catenin stabilization, driving metaplasia and tumorigenesis [127,128,129,130].

Shibata et al. found that stromal myofibroblasts expand in the gastric corpus mucosa after upregulation of SDF-1, a chemokine ligand for CXCR4, after Helicobacter infection. These studies suggested changes in myofibroblasts attendant with induction of oxyntic atrophy and metaplasia [127,131].

4.1.5. SPEM Clinical Implications

Given SPEM’s association with chronic inflammation, atrophy, and cancer, it is increasingly being considered a critical early lesion for gastric carcinogenesis. Its transient nature makes it a valuable marker for intervention. Clinical surveillance protocols may benefit from incorporating biomarker screening for SPEM, particularly in high-risk populations or in patients with atrophic gastritis or H. pylori history [132,133].

Encouragingly, SPEM can regress following H. pylori eradication. In high-risk individuals with a family history of GC, SPEM lesions diminished in nearly 70% of patients post-eradication [133]. Regression correlated with changes in miRNA profiles, including miR-21, miR-155, and miR-223, making them potential non-invasive biomarkers for surveillance.

The dual role of RSPO3—in tissue regeneration and antimicrobial defense, but also in promoting metaplasia and tumor progression—makes it a promising therapeutic target. Strategies include RSPO3/LGR inhibitors to curb excessive Wnt-driven proliferation, modulation of RSPO3 signaling to balance repair and metaplasia, and exploring interactions with YAP, Notch, and BMP pathways to fine-tune interventions [129,132]. Specific high-level surveillance guidelines for SPEM are not yet standardized. In practice, SPEM surveillance often follows the same intervals as IM due to their overlapping risk associations with gastric cancer.

4.2. Intestinal Metaplasia

Gastric IM is considered a key preneoplastic lesion in the Correa cascade [85,134], a sequential model of gastric carcinogenesis. Understanding the cellular and molecular basis of IM is crucial for developing preventive and therapeutic strategies against GC. Gastric epithelial cells, under chronic stress or injury, undergo transformations that result in the emergence of intestinal-type epithelium.

4.2.1. Global Prevalence of Gastric Intestinal Metaplasia

A comprehensive systematic review and meta-analysis including 20 studies with 57,263 individuals estimated the global prevalence of gastric IM at 17.5% (95% CI: 14.6–20.8%). The highest prevalence was reported in the Americas at 18.6% (95% CI: 13.8–24.6%), particularly among patients with gastroesophageal reflux disease, where the prevalence reached 22.9% (95% CI: 9.9–44.6%) [133].

Regional differences were substantial, with prevalence ranging from 3.4% in Northern Europe to 23.9% in South America. Ethnic disparities were also noted; for instance, Hispanics and East Asians had higher rates compared to other populations [133,135].

4.2.2. Intestinal Metaplasia Classification

Traditionally, IM has been subdivided into complete (Type I) and incomplete (Types II and III) forms based on histochemical and morphological characteristics. CIM resembles small intestinal epithelium, with the presence of absorptive enterocytes, Paneth cells, and goblet cells secreting sialomucins. In contrast, IIM mimics colonic epithelium, lacking absorptive cells and Paneth cells, and is characterized by columnar cells secreting sulfomucins. IIM is more frequently associated with dysplasia and a higher risk of progression to GC compared to the complete type [136,137,138,139].

Zheng Y. and colleagues (2010) proposed the division into two subtypes according to the atypical changes in the metaplastic epithelium: simple IM (SIM) and atypical IM (AIM). The researchers detected three tumor-associated proteins, p53, c-erbB-2, and Ki67, in different subtypes of IM in order to find which one is more associated with GC [139].

Niwa and colleagues (2005) suggested a histological reclassification of the gastric mucosa IM, emphasizing the cellular composition within metaplastic glands, rather than relying solely on mucin histochemistry [140]. This classification is based on the presence or absence of residual gastric-type epithelial cells—namely, mucous neck cells or pyloric-type cells—within metaplastic glands and has led to the identification of two distinct subtypes:

• Solely Intestinal Type (I-type)

In this type, the glandular architecture is composed exclusively of intestinal-type epithelial cells, including goblet cells, absorptive enterocytes with a brush border, and, occasionally, Paneth cells. There is no histological evidence of gastric-type cells within the metaplastic glands. This type is often associated with more advanced stages of intestinalization and typically shows strong expression of intestinal mucins MUC2 (+), along with CD10 (+) and CDX2 (+), and absence of gastric mucins, MUC 5 AC (−) [140].

• Gastric–Intestinal Mixed Type (GI-mixed type)

This type is characterized by the coexistence of gastric-type and intestinal-type epithelial cells within the same glandular unit. Specifically, goblet cells and columnar absorptive cells expressing intestinal markers are observed alongside residual gastric-type cells. Mucin histochemistry typically shows a combination of neutral and acidic mucins, and immunohistochemical staining may reveal coexpression of gastric markers such as MUC5AC (+) or MUC6 (+) with intestinal markers like MUC2 (+), CD10 (+/−). GI-mixed-type IM is considered to represent an intermediate or transitional stage in the process of metaplastic reprogramming and has been proposed to carry a higher malignant potential than solely intestinal metaplasia [140].

This classification offers a more biologically grounded understanding of gastric IM, aligning glandular phenotype with the dynamic progression of gastric epithelial reprogramming and potential neoplastic transformation.

4.2.3. IM vs. SPEM

Although IM generally presents as patchy or multifocal within the gastric mucosa, SPEM typically manifests more diffusely across the gastric corpus and fundus in patients progressing toward GC [106]. Importantly, SPEM is distinct from a mere expansion of antral-type mucosa; it lacks gastrin-producing cells and instead expresses unique markers such as HE4 (WFDC2), which are absent in the antral epithelium [141] (

Table 3. Differences in molecular markers of intestinal metaplasia (IM) and spasmolytic polypeptide-expressing metaplasia (SPEM).

Feature	IM	SPEM
Histological Features	Characterized by goblet cells, columnar epithelium, and sialomucin production (MUC2) [142]	Resembles antral glands, with MUC6 expression and absence of G-cells [142]
Mucin Expression	Mucin (MUC2) expression and trefoil factor 3 (TFF3); gastric mucins (MUC1, MUC5AC, MUC6) are downregulated in Type I [143]	Strong expression of MUC6 and trefoil factor 2 (TFF2) [71,144]
Transcription Factors	Type I: Negative for SOX2, Positive for CDX2 [142]; Types II and III: Positive for SOX2, Negative for CDX2 [145,146]	Not significantly associated with SOX2 or CDX2 expression [142,145]
Cellular Markers	Expression of CDX2, SOX2, MUC2, MUC5AC, and MUC6 in different types of IM [146]	Strong expression of TFF2, MUC6, MUC5AC, HE4 [147,148]
Mutations	C > T and T > G transitions in CpG sites; mutations in FBXW7, APC, TP53, and ARID1A are observed [149]	Not significantly linked to the same mutation profile; mostly associated with inflammation and cytokine signaling [148]
Somatic Copy Number Alterations (sCNAs)	12.5% of cases show 8q amplification (MYC oncogene) [150]	Less frequent but may be influenced by inflammatory signaling [148]
Telomere Length	Telomere shortening observed, particularly in antral IM [151]	Not widely observed in SPEM but may be influenced by inflammatory factors [151]
DNA Methylation	Increased DNA methylation, particularly in antral IM [152]	DNA methylation is less studied but could be involved in inflammatory pathways [149]
miRNA Expression	Upregulation of miRNA-146a, miRNA-155, miRNA-17-92 cluster; downregulation of miRNA-490-3p, miRNA-30a [153]	Not extensively studied for specific miRNA changes; inflammatory miRNAs (e.g., miRNA-21) may be relevant [153]
Progression to GC	IM, particularly incomplete IM (Type II and III), is associated with an increased risk of GC [154]	No direct role in GC progression but associated with inflammation and epithelial changes [148]
SPEM and Inflammatory Pathways	Occurs in response to chronic injury, H. pylori infection, or acid reflux [114]	Strong involvement of M2 macrophages and inflammatory cytokines such as IL-6 and IL-1β [114,115]

). Despite growing evidence linking these metaplastic lineages to neoplastic transformation, the precise relationship between IM and SPEM, as well as their respective roles in the development of GC, remains incompletely understood.

4.2.4. Gastric Cancer Risk Associated with Intestinal Metaplasia

Patients with IM in the corpus or those with IIM have a higher risk of GC compared with patients demonstrating IM confined to the antrum or CIM [155,156,157,158].

A meta-analysis of 21 studies, encompassing over 400,000 participants, reported that individuals with IM had a pooled odds ratio (OR) of 3.58 (95% CI: 2.71–4.73) for developing GC compared to those without IM. This risk was particularly pronounced for IIM (pooled OR = 9.48; 95% CI: 4.33–20.78), while CIM had a modest association (pooled OR = 1.55; 95% CI: 0.91–2.65) [157].

IIM is associated with a markedly increased risk of progression to GC. A meta-analysis of 12 studies found that patients with IIM had a pooled relative risk (RR) of 4.48 (95% CI: 2.50–8.03) compared to those with CIM. Within the IIM category, Type III IM exhibited the highest risk, with a pooled RR of 6.27 (95% CI: 1.89–20.77) when compared to types I and II [159].

Another meta-analysis focused on cohort studies showed that patients with IIM had a pooled RR of 5.16 (95% CI: 3.28–8.12) for GC compared to those with complete IM. Moreover, Type III IM showed a significantly higher risk than Type II IM (pooled RR = 2.88, 95% CI: 1.37–6.04). Additionally, another meta-analysis found the incidence rate of GC in patients with IM to be 3.38 per 1,000 person-years (95% CI: 2.13–4.85), and the progression to dysplasia was 12.51 per 1,000 person-years (95% CI: 5.45–22.03) [156].

The predictive value of IIM for GC or dysplasia remains controversial. Both the British Society of Gastroenterology and the MAPS II guidelines emphasize the need for more robust evidence before incorporating IIM into clinical decision-making [159,160,161,162]. A meta-analysis by Wei et al. demonstrated that patients with IIM had significantly increased risks of GC (RR = 4.96; 95% CI: 2.72–9.04), dysplasia (RR = 4.82; 95% CI: 1.45–16.0), and combined cancer/dysplasia outcomes (RR = 4.48; 95% CI: 2.50–8.03) compared to those with complete IM (CIM) [159,161].

4.2.5. IM Mechanisms of Development and Biomarkers

The development of IM is typically preceded by chronic gastritis and atrophic alterations in the gastric mucosa. Several key factors contribute to this process:

• H. pylori infection: Persistent infection promotes chronic inflammation, oxidative stress, and cytokine release, all of which contribute to mucosal injury and remodeling [132].
• Loss of parietal cells: Parietal cell depletion results in glandular atrophy and significant changes in the gastric microenvironment, facilitating metaplastic transformation [82].
• Transdifferentiation and dedifferentiation: Lineage plasticity among differentiated cells, such as chief cells and mucous neck cells, allows for their conversion into metaplastic cell types [52].
• Stem cell reprogramming: Aberrant activation or reprogramming of gastric stem cells in response to chronic injury may initiate intestinal differentiation [163].

Multiple molecular pathways and transcriptional regulators are implicated in IM pathogenesis:

• CDX2: This intestine-specific transcription factor is central to driving intestinal-type differentiation in gastric epithelial cells [164].
• Notch, Wnt/β-catenin, and Hedgehog signaling: These developmental pathways regulate proliferation, differentiation, and cell fate decisions during metaplastic progression [165].
• Epigenetic modifications: Aberrant DNA methylation and histone acetylation patterns are frequently observed in IM and may promote or stabilize the metaplastic phenotype [152].

Cellular Origins of IM

• Chief cell transdifferentiation (SPEM): Studies suggest that gastric chief cells can transdifferentiate into spasmolytic polypeptide-expressing metaplasia (SPEM), which may serve as a precursor stage to IM [52].
• Mucous neck cell lineage: Under chronic inflammatory conditions, mucous neck cells may also undergo reprogramming and contribute to the metaplastic cell pool [132].
• Stem cell plasticity: Chronic injury and altered signaling cues can redirect gastric stem cells toward an intestinal lineage, highlighting the role of stem cell plasticity in metaplasia development [28,163].

KLF4 (Kruppel-like factor 4) and OLFM4 (Olfactomedin 4) have recently emerged as potential biomarkers and functional mediators in IM of the stomach. KLF4 is strongly upregulated in IM, suggesting a central role in this phenotypic transformation. KLF4 is minimally expressed in normal gastric epithelium but is significantly increased in areas of IM. Studies have demonstrated colocalization of KLF4 with goblet cell markers such as MUC2 and transcription factors like CDX2 in metaplastic gastric glands [166,167]. KLF4 promotes intestinal-type differentiation in the stomach by activating transcription of genes such as MUC2 and TFF3, contributing to the establishment of an intestinal phenotype [166]. It also works synergistically with CDX2, a key inducer of intestinal identity, to drive the reprogramming of gastric epithelial cells. KLF4 expression is regulated by inflammatory stimuli and epigenetic mechanisms, which are often triggered by chronic H. pylori infection—a major risk factor for IM. Moreover, KLF4 is a downstream effector of the Wnt/β-catenin and Notch pathways, which are both involved in epithelial lineage specification [167,168].

OLFM4, also known as Olfactomedin-4, GW112, or hGC-1, is a glycoprotein belonging to the olfactory regulatory protein family. OLFM4 expression is absent in normal gastric mucosa but is present in the small intestine and colon. OLFM4, in combination with Myosin heavy chain 9 (MYH9), accelerated the ubiquitination of GSK3β and resulted in increased β-catenin levels through the Wnt signaling pathway, promoting the proliferation and invasion abilities of PLGC cells. Wei et al. showed that OLFM4 expression is highly increased in IIM, with superior diagnostic accuracy of IIM when compared to CDX2 and MUC2. OLFM4, along with MYH9, was overexpressed in IM organoids and PLGC animal models [169].

The effects of H. pylori on the gastric epithelium have been extensively studied, with one of the most important pathogenic factors being cytotoxin-associated gene A protein (CagA) positive strains. Virtually all of East Asian strains and 60% of Western strains of H. pylori are cagA+, with infected patients developing more distinct inflammation, gastric ulceration and higher risk of GC [170,171]. Bacterial CagA protein interacts with a series of host epithelial proteins, including ASPP2, RUNX3, PI3K, SHP2 and E-cadherin, resulting in the degradation and inactivation of p53 and RUNX3, deregulation of the PI3K-AKT, Ras-ERK and Wnt pathways, and disruption of adherens junctions [172]. CagA has also been shown to alter DNA methylation patterns, further deregulating normal epithelial gene expression patterns. IM samples show higher levels of methylation than atrophic gastritis samples, suggesting that DNA methylation pattern changes may play a vital role in the Correa model of intestinal GC [173,174].

4.2.6. Genomic and Epigenomic Alterations in IM

IM harbors distinct mutational and epigenomic features that may contribute to its progression toward GC. Mutational signatures in IM are characterized predominantly by C > T transitions at CpG sites, a pattern associated with age-related mutagenesis. In addition, T > G mutations, possibly linked to gastric acid exposure, have also been observed [175,176]. While TP53 and ARID1A are among the most frequently mutated genes in GC, they are not commonly altered in IM. Notably, FBXW7 was the only significantly mutated gene in IM detected by MutSigCV analysis, but its mutation rate remains low (4.7%) compared to GC (up to 18.5%), indicating that additional oncogenic events are necessary for malignant progression. APC mutations have also been sporadically reported in IM cases [175,177]. A recent study of genomic and epigenomic profiling of IM showed that IM has a low mutational burden compared to non-hypermutated GC (2.6 vs. 6.9 mutations/Mb) and harbors recurrent mutations in certain tumor suppressor genes like FBXW7. DNA methylation profiling showed that the majority of IM patients in the high methylation group had relatively high mutational load, frequent chromosomal copy number variations, and FBXW7 mutations, which occurred mainly in the antrum [178].

Somatic copy number alterations (sCNAs) are present in approximately 12.5% of IM cases, with amplification of chromosome 8q22.3–8q24.3 being the most frequent event. This region includes the MYC oncogene, which may contribute to oncogenic signaling. Additionally, IM exhibits telomere shortening, particularly in antral IM compared to corpus or cardiac IM. A correlation between 8q amplification and telomere attrition has been proposed, suggesting a potential mechanistic link [175].

Epigenetically, IM displays increased DNA methylation relative to chronic gastritis. Antral IM tends to be more hypermethylated than corpus/cardiac IM, and elevated methylation levels correlate with higher mutation burden and sCNAs [175,178]. In contrast, global hypomethylation, often seen in more advanced lesions, is typically absent in IM, indicating it is a later event in gastric carcinogenesis. Hypermethylation of HOXA5 has also been implicated in the pathogenesis of cardiac IM [179]. Other molecular changes include loss of RARβ expression, abnormal CD44 transcript variants, and the presence of microsatellite instability [176].

IM is associated with distinct microRNA (miRNA) expression patterns, which may serve as biomarkers or functional contributors to pathogenesis. Elevated levels of miR-146a and miR-155 have been reported in H. pylori-associated IM, suggesting a role in inflammation-mediated epithelial changes [180]. Similarly, the miR-17–92 cluster (including miR-17-5p, -20a, -18a, -19a/b, and -92a) is upregulated in IM and may represent a serum biomarker for early detection of IM and GC [181].

Conversely, downregulated miRNAs in IM include miR-490-3p and miR-30a, both of which have been implicated in the regulation of chromatin remodeling and cell adhesion, respectively [182,183]. Additional dysregulated miRNAs include miR-486-5p, miR-645, miR-624, miR-504, and miR-106b, highlighting a broader miRNA network involved in metaplastic transformation [184].

Collectively, these findings support a model in which progression of IM toward GC is driven by the accumulation of genetic mutations, sCNAs, telomere erosion, DNA methylation, and miRNA dysregulation [175].

Hox and ParaHox genes, especially CDX1, CDX2, and Pdx1, are essential regulators of gut development and intestinal differentiation. CDX1 and CDX2, members of the Caudal-related homeobox family, play pivotal roles in maintaining the intestinal epithelial phenotype and are aberrantly expressed in gastric IM and associated lesions [185,186]. CDX2, in particular, is an early marker and potential initiator of IM, with its ectopic expression linked to H. pylori infection, chronic gastritis, and dietary factors [187,188]. Transgenic mouse models show that CDX1 drives full intestinal-type metaplasia, while CDX2 promotes PPM [189]. Both genes regulate key intestinal markers (e.g., MUC2, TFF3, SI, CA1), and their expression is modulated by Wnt, retinoic acid (RA), FGF, and MAPK signaling pathways [190]. CDX2 also induces TFF3, a trefoil factor expressed in goblet cells and IM, while repressing gastric markers like TFF1/2, indicating a central role in the intestinalization of gastric mucosa [187].

The ParaHox gene family, particularly pancreatic and duodenal homeobox 1 (Pdx1), plays a pivotal role in gastrointestinal development and metaplasia. Pdx1, also known by several aliases (IPF-1, STF-1, IDX-1), is critical for the development of the pancreas, duodenum, and antropyloric mucosa. It is expressed in hyperplastic endocrine cells during CAG and is implicated in the maturation of gastrin-producing cells. Its expression is not required for gastric amylin expression but is influenced by various stimuli, including gastrin, cytokines, and microbial factors [191]. Pdx1 is not typically expressed in SPEM, but is found in PPM and IM, particularly co-expressed with MUC6, suggesting a potential role for Pdx1 in the transition from PPM to IM [192].

The Hedgehog signaling pathway, mainly through Sonic Hedgehog (Shh) and Indian Hedgehog (Ihh), regulates gut patterning, epithelial differentiation, and mesenchymal–epithelial interactions. Shh is highly expressed in the stomach and duodenum and is essential for stomach patterning, while Ihh influences stem cell proliferation [193].

• Loss of Shh is associated with gastric atrophy and IM, especially CIM, marking it as an early event in gastric carcinogenesis [194].
• Inflammatory states upregulate Shh expression, potentially contributing to stem cell regeneration but also posing a risk for neoplastic transformation [194].
• Downstream effectors like Gli3 and mesenchymal targets such as Bmp4 and Foxf integrate Hh signals with Wnt signaling, modulating proliferation, apoptosis, and differentiation [194].

The Sox (Sry-related HMG box) family, especially Sox2, contributes to gastric identity and opposes intestinal differentiation. Sox2 is highly expressed in the rostral (gastric) epithelium and is downregulated toward the duodenum, where Cdx genes dominate. In IM and GC, Sox2 is downregulated, whereas Cdx1/2 are upregulated, supporting a gastric-to-intestinal transdifferentiation model [195].

Gastric mesenchyme induces Sox2 and suppresses Cdx2, while intestinal mesenchyme has the opposite effect, highlighting region-specific epithelial patterning through mesenchymal–epithelial interactions [196]. These interactions underscore the importance of epithelial–mesenchymal signaling in directing cell fate decisions and metaplasia in the gastrointestinal tract [197].

The coordinated action of Pdx1, Hedgehog, and Sox genes reflects a complex interplay between intrinsic transcriptional regulators and extrinsic signals from the microenvironment. These pathways collectively influence gastric epithelial plasticity, the development of metaplasias such as PPM and IM, and potentially the early events of gastric carcinogenesis.

POU proteins are critical transcription factors expressed during early embryogenesis and are essential for organ development and cell differentiation. They contain a conserved DNA-binding domain composed of a POU-specific domain and a homeodomain. Oct-1 and Oct-2 are class II POU proteins that recognize a conserved octamer motif, while Pit-1, a class I member, has a different binding sequence and regulates pituitary hormones.

Oct-1 has been implicated in regulating Cdx2, a key transcription factor in intestinal identity. It binds to the Cdx2 promoter and may induce its expression in the gastric mucosa, contributing to the development of IM [185]. Oct-1 expression is elevated in IM and intestinal-type GCs, suggesting a role in gastric carcinogenesis, potentially through Cdx2 regulation [185,187].

RUNX3, a tumor suppressor, is normally expressed in gastric chief cells, but is frequently silenced in IM and GCs due to promoter methylation or gene deletion. Loss of RUNX3 is associated with GC progression, possibly allowing ectopic Cdx2 expression and promoting intestinal differentiation [198].

The MAPK/ERK pathway, particularly ERK1/2, is activated by H. pylori infection and plays a key role in gastric epithelial proliferation and altered apoptosis, contributing to the cascade from IM to dysplasia and eventually GC. Activation of MAPK is also associated with early intestinal differentiation markers such as villin, with H. pylori triggering transcriptional activation involving Elk-1 and SRF [199].

Potential molecular biomarkers for IM progression and gastric carcinogenesis include aneuploidy, overexpression of cyclin D1 and growth factors, and loss of heterozygosity at p53 and p16 [200,201].

4.2.7. IM Clinical Implications and Management Strategies

Given the increased risk of GC associated with gastric IM, particularly the incomplete subtype, clinical guidelines recommend risk stratification and surveillance. Patients with incomplete or extensive gastric IM, especially those with additional risk factors such as a family history of, may benefit from regular endoscopic monitoring. The European Society of Gastrointestinal Endoscopy (ESGE), European Helicobacter and Microbiota Study Group, and the European Society of Pathology recently updated their consensus in 2025. They recommend that patients with extensive precancerous changes—such as advanced IM or high OLGIM stages (III/IV)—undergo surveillance endoscopy every 3 years, regardless of etiology [162]. In contrast, patients with minimal or antral-restricted IM without additional risk factors generally do not require surveillance [162]. For patients with mild to moderate atrophy restricted to the antrum, there is no evidence to recommend surveillance [202]. Atrophy of the gastric mucosa is defined as the decrease or disappearance of the original gastric glands, which may be replaced by SPEM/PPM, pyloric metaplasia, IM, or fibrosis [202]. IM observed during endoscopy is considered the most reliable indicator of gastric atrophy. Patients exhibiting advanced gastritis—specifically, atrophy and/or IM involving both the antrum and corpus—are regarded as having a high risk of developing GC [162]. High-definition endoscopy combined with chromoendoscopy or image-enhanced endoscopy using magnifying endoscopy (ME), such as narrow-band imaging (NBI), offers superior diagnostic accuracy compared to conventional white-light endoscopy [162,203]. Additionally, artificial intelligence has been increasingly utilized for diagnosing CAG using various imaging modalities. For instance, convolutional neural network (CNN) models trained on standard white-light images have demonstrated diagnostic accuracies exceeding 90% (93–94.2%), outperforming expert endoscopists and effectively differentiating H. pylori-associated gastritis from autoimmune gastritis [204]. Furthermore, when distinguishing CAG from GC using ME-NBI images, CNN models achieved an accuracy of 85.3%, with a sensitivity of 95.4% and specificity of 71.0%, processing each image in just 0.02 s [202,205].

A 2024 systematic review and meta-analysis of AI-assisted endoscopy for IM reported sensitivity of 94%, specificity of 93%, and the area under the curve (AUC) of 0.97—significantly outperforming human endoscopists (95% vs. 79% sensitivity) [4]. While these findings are promising, the studies were mostly retrospective, and prospective validation in real-world practice remains a critical next step [206]. Similarly, a meta-analysis of AI-assisted diagnosis of early gastric cancer showed an accuracy of 90.32% for diagnosing EGC, significantly superior to that of senior endoscopists (70.16% ± 8.78%) [207]. However, most studies were retrospective, single-center, and technically heterogeneous, highlighting the importance of prospective, multicenter trials [208].

5. Discussion

A change in identity can be achieved via differentiation of less specialized cells or via dedifferentiation or transdifferentiation of specialized cells. These phenomena are grouped under the umbrella term of ‘‘cellular plasticity’’. The stomach lining exhibits significant cell plasticity, meaning differentiated cells can reprogram into other cell types, including stem cells. This plasticity allows the stomach to repair itself after injury, but it can also lead to cycles of differentiation and dedifferentiation, increasing the risk of accumulating mutations. Differentiated mature cells can be reprogrammed into another mature cell as a reserve stem cell to replace lost cells. Given the importance of cell fate dynamics in precancerous lesions, there is substantial interest in understanding the mechanisms that govern cell plasticity. Many aspects of stomach specification are still poorly understood or unexplored [28]. While it is evident that interactions between the gastric endoderm and mesoderm play a crucial role in gastric development, little is known about the mechanisms by which naive endodermal progenitors are directed to become gastric progenitors.

Over the past decade, numerous examples have emerged to support plasticity in differentiated cells. First, it became clear that normal, somatic cells could be reprogrammed to pluripotency [26]. Furthermore, in tissues, injury can induce a repair process that recruits largely post-mitotic, differentiated cells back into the cell cycle in most, if not all, organs and species, for example, glia [116,209], lung [210], and heart [211], in multiple gastrointestinal tracts [24]. Each such example to date has been studied essentially in isolation within the context of a particular type of injury and a single organ; however, because the process is so widespread, we have postulated that it may be governed by a shared, evolutionarily conserved molecular and cellular program that is independent of tissue and species [24]. This cellular program occurs during cell fate changes in various types (e.g., reversion, dedifferentiation, transdifferentiation, reprogramming). The lack of a standard term for the actual cellular process itself impedes finding shared features that transcend cell types, tissues, and model systems. Willet S. et al. proposed a new, unifying term: “paligenosis” from the Greek: pali/n/m (meaning backward or recurrence) + genea (born of, producing) + osis (an action or process) [114]. But not all differentiated cells are likely to be able to undergo paligenosis. In the stomach, for example, Huh et al. have never observed this phenomenon in mature parietal cells [24,82]. Cells that are constitutively undifferentiated and replicative, like those of the isthmus of the stomach or LGR5+ crypt-base columnar cells, should not need any stage of paligenosis. They may acquire the building block nucleotides and amino acids from the blood and/or extracellular environment, given that, by definition, their lack of differentiation means they contain limited non-nuclear components to recycle. Other cells, such as mucous neck cells in the stomach or +4 cells in the intestine [212], may be able to respond to injury but are less well differentiated and thus may be able to skip the autodegradative phase and go directly to the activating mTORC1 and cell division phase of paligenosis.

Cellular reprogramming, whether by dedifferentiation or transdifferentiation, requires dramatic changes in gene expression. Such changes are needed to activate genes associated with the new fate and to inhibit genes associated with the previous fate. In this regard, it is worth noting that transdifferentiation need not be an “all or nothing” phenomenon. For example, cellular phenotype may change as a result of transcriptional or post-transcriptional changes, such as overexpression of the YAP transcription factor in the dedifferentiation of secretory cells in the lung, but then revert back in the absence of subsequent epigenetic modifications once the stimulus is removed. Another case in point is hepatocyte transdifferentiation, where some transdifferentiating cells revert to the hepatocyte fate when the injury is resolved. It seems theoretically likely that cells that have undergone stable transdifferentiation are maintained in their new (biliary) identity, while cells that are in the midst of the transdifferentiation process can return to their original hepatocyte identity [4]. During reprogramming, cells not only undergo transcriptional changes but also exhibit epigenetic changes DNA methylation and histone modifications [212].

Metaplasia describes changes within tissues whereby one cell type is replaced with another [201]. Although the phenomenon may involve transdifferentiation, it has also been proposed that metaplasia can also result from cell migration or altered cell fate decisions during normal differentiation—so-called “transfating”. Tissue metaplasia is frequently associated with a predisposition to cancer [96]. One of the most common examples of metaplasia is “Barrett’s esophagus,” in which the normal squamous epithelium of the esophagus is replaced by columnar cells with features of the intestinal epithelium. Barrett’s metaplasia is strongly associated with the development of esophageal adenocarcinoma [213,214], which has led to the notion that transdifferentiation is associated with oncogenic risk. However, the extent to which the Barrett’s-associated risk of cancer is due to transdifferentiation remains unclear. Whether or not Barrett’s is a true metaplastic event, uncontrolled cell plasticity has been hypothesized to be carcinogenic.

The development of IM is a multifaceted process. L. Gutiérrez-González and N.A. Wright explored its molecular characteristics, highlighting the upregulation of genes such as Cdx1, Cdx2, Pdx1, Oct1, and TFF3, alongside the suppression of Hedgehog signaling. Additionally, Runx3 is silenced through epigenetic mechanisms, while pathways including Wnt and MAPK/ERK are implicated. These molecular alterations provide insight into the mechanisms underlying IM and emphasize the importance of gene regulation in the pathogenesis of GC [201].

Notch signaling plays a pivotal role in directing the fate of gastric epithelial cells, particularly in the context of injury and regeneration. Activation of the Notch pathway can induce proliferative, undifferentiated gastric progenitor cells to adopt specific lineage outcomes. For example, enhanced Notch signaling has been shown to inhibit the differentiation of secretory lineages, such as mucous neck cells and chief cells, and instead promotes lineage conversion toward absorptive or intestinal-like phenotypes, contributing to IM development. This transformation is mediated through Notch-dependent transcriptional regulators, including Hes1, which suppresses the expression of genes necessary for terminal differentiation. In chronic injury settings, persistent Notch activation can therefore promote cellular reprogramming and metaplastic remodeling, steering gastric epithelial cells toward a proliferative, metaplastic state that resembles intestinal epithelium—an early hallmark in the Correa cascade of gastric carcinogenesis [33,215].

These findings suggest that the signals that allow adult cells to change their identity may put tissues at increased risk of malignant transformation. Such an association could provide an additional explanation for the well-appreciated association between tissue inflammation following chronic injury (the in vivo stimulus for transdifferentiation) and cancer.

At the cellular level, changes in identity can occur experimentally through the activation of specific signaling pathways or physiologically as a response to signals released during injury and inflammation. These processes are tightly regulated, as uncontrolled plasticity could destabilize tissues and/or lead to cancer. Although most examples of plasticity discussed in this review were conducted in model organisms, it is likely that the cellular relationships identified in these studies are conserved in humans. For example, one study found that human hepatocytes transplanted into mouse livers undergo transdifferentiation into biliary cells, indicating that human hepatocytes also undergo transdifferentiation as a physiological response to injury. Generating new β cells in the pancreas as a treatment for diabetes has high priority in the field of regenerative medicine, and transdifferentiation—from either α cells or other endoderm derivatives—is an attractive approach [216].

Goblet cell metaplasia, which can also be observed in the ciliated bronchial mucosa, should be discriminated from true IM because it lacks intestinal cell lineage and may more properly be named goblet cell hyperplasia, a process provoked by inflammatory cytokine signaling [217]. Another example is the “pseudogoblet cells” observed in the esophagus, whose presence can be misinterpreted as evidence of Barrett’s esophagus. Because these changes are also not accompanied by other intestinal cell lineages, these examples may be better defined as transdifferentiation rather than metaplasia. In contrast, metaplasia should be defined as a phenotypic conversion through reprogramming of differentiation at the level of stem/progenitor cells, thereby resulting in clonal changes in the whole gland. “Transcommitment” may be a preferred term for the reprogramming event causing clonal changes in the gland, rather than the widely misused “metaplasia,” to avoid further misunderstanding. In contrast, metaplasia should be defined as a phenotypic conversion through reprogramming of differentiation at the level of stem/progenitor cells, thereby resulting in clonal changes in the whole gland [70].

Under a strict definition of metaplasia as a transcommitment process—where stem or progenitor cells undergo genetic reprogramming to adopt a new lineage identity—the designation of SPEM observed following acute parietal cell loss induced by DMP-777 may be inaccurate. This is because it does not appear to be a clonal reprogramming event that shifts the gastric gland toward an intestinal phenotype. Additionally, cells expressing spasmolytic polypeptide (also known as trefoil factor 2), typically found in the isthmus of normal gastric glands, may relocate toward the gland base following parietal cell depletion. The SPEM-like changes induced by DMP-777 are reversible once the drug is withdrawn, indicating that they likely do not originate from a reprogramming of stem or progenitor cells. In atrophic human corpus mucosa, pyloric-like glandular changes are often seen. Wada et al. reported the occurrence of both “pyloric gland metaplasia” and “pseudopyloric gland metaplasia” in the oxyntic mucosa, which disappeared within six years after H. pylori eradication—suggesting these alterations are temporary adaptive responses rather than stable phenotypic conversions [218]. Moreover, spasmolytic polypeptide-expressing cells located at the base of human IIM glands should not be classified as SPEM but rather as part of the IIM cellular architecture [70].

Both corpus and antral epithelia exhibit notable plasticity. Chief cells in the corpus can dedifferentiate and contribute to regeneration, particularly during injury-induced metaplasia [40]. In the antrum, reserve stem cells replace lost Lgr5+ cells, ensuring continuity of epithelial renewal [58]. For example, tamoxifen administration can induce parietal cell loss and provoke metaplastic changes, obscuring natural lineage behavior [98]. Heterogeneity in mouse SPEM lineages identifies markers of metaplastic progression [35,86].

Lee SH and colleagues identified distinct metaplastic gland phenotypes by employing immunostaining markers targeting different epithelial lineages. Although normal corpus glands do not express AQP5, their analysis demonstrated strong apical AQP5 expression in SPEM cell populations. Additionally, they noted glandular formations with AQP5-positive cells at the base and Trop2 and TFF3 expression in the upper segments—characteristics consistent with IIM. Trop2 expression was elevated in IIM but not observed in CIM. Similarly, AQP5 was absent in CIM glands; however, CD44v9-positive cells remained detectable at the gland bases. These results suggest that SPEM cells are frequently present at the base of glands undergoing IIM [73].

Notably, AQP5 and GSII double-positive SPEM cells were identified at the base of IM and dysplastic glands in Mist1-Kras mice at 3 and 4 months following Kras induction. These glands also exhibited increased luminal Trop2 expression, indicating that the observed metaplasia in this model aligns with an IIM phenotype [73].

SPEM expresses markers such as TFF2, MUC6, HE4, CD44v9, and AQP5 and frequently occurs adjacent to dysplasia or adenocarcinoma [219]. The expression of these markers supports the clonal expansion hypothesis, whereby SPEM acts as a transitional state toward IM and GC. In the gastric corpus, SPEM morphologically resembles the glandular architecture of the antral mucosa (“antralization”). This condition corresponds to the clinical term “atrophic gastritis,” referring to the loss of parietal and chief cells (atrophy) and the accompanying chronic inflammation (gastritis) that drives the pathological transformation [24]. Such changes are potentially reversible if the injurious agent is removed [95,219]. However, in the presence of sustained damage, SPEM becomes irreversible and may progress to IM, initiating the Correa cascade [89,138].

Hayakawa et al. proposed two possible pathways for the development of dysplasia and GC. In one, low Notch signaling allows stem cell activation followed by differentiation into either SPEM or IM, with possible interconversion between these states and progression toward dysplasia or carcinoma marked by TFF2 expression. In the alternative pathway, high Notch activation in stem cells directly promotes oncogenic transformation without intermediate metaplasia [93,94].

Two competing hypotheses exist regarding the cellular origin of SPEM and related metaplastic lineages. The first, dominant hypothesis posits that mature chief (zymogenic) cells are the source, based on evidence that chief cells expressing Mist1 and Lgr5 undergo transdifferentiation into metaplastic lineages [27,52,220]. A landmark multicenter study in 2010 confirmed this by demonstrating a progressive transformation of chief cells in human gastrectomy specimens, consistent with prior murine data [52]. These transitioning chief cells, which normally express Mist1, acquire TFF2 in SPEM or CDX2 in IM, forming an intermediate lineage. Mills and Goldenring articulated four lines of evidence supporting chief cells as the origin of metaplasia [221], and they argue that misinterpretations of foveolar hyperplasia as true metaplasia have led to confusion in the literature. Other studies based on β-galactosidase lineage tracing also support a chief cell origin but require further corroboration. The alternative hypothesis suggests that isthmal progenitor cells may also contribute to metaplasia [222]. Hayakawa et al. demonstrated, using mouse models, that Mist1+ isthmal cells could give rise to SPEM, as ablation of these cells disrupted lineage tracing, while ablation of Mist1+ chief cells did not [222]. Further research by Kinoshita’s group indicated that mature chief cells were not the principal contributors to SPEM; instead, precursor cells located in the isthmus served as the cellular source of persistent metaplasia [93,94].

In 2022, Mills and Goldenring proposed a unifying model in which both mature chief cells and isthmal progenitors contribute to SPEM through different mechanisms: transdifferentiation of Mist1+ chief cells and downward migration of proliferating isthmal cells mimicking a pyloric phenotype [221]. They introduced a new subclassification—pyloric metaplasia with foveolar hyperplasia—characterized immunohistochemically as MUC6+, TFF2+, and PGI−.

Thus, two principal sources of metaplastic lineages are now recognized: long-lived chief cells in the basal gastric glands undergoing paligenosis and isthmal progenitors responding to inflammation via proliferation and expansion into glandular compartments. Whether isthmal cell proliferation represents true metaplasia or foveolar hyperplasia remains debated, especially since TFF2 expression is also a feature of normal isthmal reserve cells.

Gastric mucosal metaplasia with an antral-like phenotype is now viewed as an early, and potentially reversible, precursor to IM—particularly the incomplete type—and is widely considered an early event in the gastric carcinogenesis sequence. The metaplastic lineage termed SPEM, located at the base of metaplastic glands, expresses spasmolytic polypeptide (TFF2) and shows reparative capacity. The existence of a transitional metaplastic phenotype retaining partial zymogenic functions remains under investigation, and its reversibility is uncertain. More comprehensive studies are needed to clarify the link between pyloric-type metaplasia and intestinal-type gastric adenocarcinoma.

Although a number of cell lineage markers including human epididymis protein 4 (HE4), TFF2, MUC6, or CD44v9 for SPEM and TFF3, MUC2, caudal type homeobox (CDX)1/2, or alpha defensing 5 (DEFA5) for IM have been considered to determine the identities of metaplasias, it is not clear how the various glandular components are distributed in different types of metaplastic glands and whether any lineage-specific markers or marker combinations can more definitively distinguish the metaplastic gland types that may be associated with a higher risk of GC development. The question of whether SPEM and IM warrant reclassification is an ongoing discussion in the field of GC research. While both SPEM and IM are associated with GC development, their precise roles and relationship to each other and to cancer initiation are still being investigated [106]. While SPEM and IM are closely related and both associated with GC, it is not yet clear if they should be reclassified as a single entity. Further research is needed to fully understand their roles in gastric carcinogenesis and to determine whether a reclassification is warranted [106].

The molecular mechanisms leading to IM induction are not fully elucidated. There is evidence for both direct bacterial action [188] and indirect mechanisms through inflammatory cytokine expression acting to increase CDX2 expression. Other, non-H. pylori gastric bacteria may also participate in this process by stimulating inflammation [223]. Chronic H. pylori infection also promotes methylation of CpG sites in chronic gastritis, but during the progression to IM and on to GC, there is progressive demethylation of the CDX2 gene promoter, resulting in higher CDX2 expression [186,187]. Specific microRNAs (miRs) may also play a role in this aberrant expression of CDX2. Interestingly, several other factors considered injurious to the gastric mucosa, such as bile acids [69], can also contribute to inducing CDX gene expression, leading to gastric IM development, and they may also promote progression to neoplasia. The molecular events leading to gastric IM, however, have not been studied in the context of autoimmune gastritis.

IM is believed to predispose to GC, although this association is not fully accepted. Epidemiological studies have indicated that compared to individuals without, those with IM have a 10-fold increased risk of developing GC, with Type III IM carrying the greatest risk [159,160]. Some studies have also shown that infection with H. pylori is one of the major aetiological factors contributing to the development of IM and progression through to GC [70,80,133]. Persistent gastric mucosal irritation caused by H. pylori infection leads to IM that is believed to arise due to the differentiation of gastric stem cells towards cells of an intestinal cell type rather than becoming cells of a gastric phenotype.

A variety of signaling pathways are essential for controlling gastric epithelial cell plasticity, working in coordination with multiple regulatory factors through intricate interactions and feedback mechanisms. The key signaling pathways that drive phenotypic transformation of gastric epithelial cells include the following: Wnt signaling pathway, Notch signaling pathway, Hedgehog (Hh) signaling pathway, transforming growth factor-β (TGF-β) signaling pathway, fibroblast growth factor (FGF) signaling pathway, and Hippo/YAP signaling pathway [4].

There are still gaps in our understanding of cellular dynamics, particularly a lack of knowledge about the dynamic processes of cell state transitions at the single-cell level. The relationship between epithelial and mesenchymal states and stem cell states remains unclear. While single-cell RNA sequencing offers detailed insights into cellular diversity, it faces technical challenges—such as potential loss of certain cell types during sample preparation—that can hinder the precise identification and analysis of cell subpopulations [224]. Additionally, existing in vitro and in vivo models fall short in replicating the dynamic and spatial aspects of gastric epithelial cell plasticity. Although spatial transcriptomics holds promise for exploring how cellular phenotype changes during disease progression influence interactions between stromal and epithelial components, its use in GC research is still at an early stage [4].

The specific upregulation of KLF4 and OLFM4 in IM makes it are valuable biomarkers for histological diagnosis and potentially for predicting the progression to gastric adenocarcinoma. Their expression pattern in IM correlates with other intestinal markers and may enhance diagnostic accuracy in early gastric lesions [168].

6. Future Research Directions

Advancements in our understanding of cellular plasticity will undoubtedly enrich our knowledge of stem cells and differentiated cells, particularly their roles in tissue repair. Such insights will also shed light on the molecular mechanisms that underlie these cellular processes, offering refined models for tumorogenesis. Further research is necessary to identify which cell types possess stem cell capabilities in the adult stomach—especially within the corpus and to clarify their relative roles during normal maintenance as well as in response to injury or disease [28].

While many cancers are thought to arise from stem cells, in the stomach, chief cells play a dual role—acting both as direct contributors to tumor formation and as sources of stem or progenitor cells that may initiate malignancy. Nevertheless, the precise mechanisms by which chief cells and stem cells drive metaplasia and GC remain poorly understood. Future investigations are likely to focus on the interactions between epithelial and stromal cells that shape the stem cell niche, as well as on the role of immune system modulation.

Advanced tools such as single-cell genomics, human organoid systems, and humanized animal models hold great promise for deepening our insight into the complex processes underlying gastric cancer development. Additionally, the extracellular matrix, a key component of the adult stem cell niche, remains insufficiently characterized—particularly in how its molecular composition, physical properties, and receptor interactions influence stem cell dysfunction during gastric tumorogenesis. These mutations play a pivotal role in dictating cellular plasticity, influencing the adaptive responses of cancer cells to their microenvironment, enabling cancer cells to dynamically adapt, evade therapeutic interventions, and manifest resistance. Deciphering the intricate interplay between driver mutations and cellular plasticity in metaplasia is imperative for the formulation of targeted therapeutic approaches that address the heterogeneity inherent in this malignancy, thereby advancing precision medicine and improving patient outcomes [225].

Understanding the molecular basis of cellular plasticity in the gastric epithelium remains a critical challenge in gastroenterology and cancer biology. Although recent studies have identified key transcription factors and signaling pathways—such as Notch, WNT, and IL-13—as well as progenitor-like cellular states associated with metaplastic remodeling, the precise temporal and hierarchical relationships among these factors remain poorly delineated [40,116,225]. Future research should aim to

• Elucidate Early Triggers: Investigate how acute and chronic injuries, such as H. pylori infection, initiate reprogramming events at both the cellular and chromatin levels [226].
• Resolve Cellular Hierarchies: Apply single-cell multi-omics and lineage-tracing technologies to define the precise origins, trajectories, and fates of metaplastic cells, particularly in the transition from chief cells to SPEM and ultimately to IM [227,228].
• Decode Stromal–Epithelial Crosstalk: Characterize the impact of immune signaling (e.g., IL-13, IL-33, IL-25), fibroblasts, and mesenchymal populations in modulating epithelial plasticity and driving metaplastic progression [229,230].
• Identify Epigenetic Switches: Explore the function of chromatin remodelers, non-coding RNAs, and DNA methylation in priming and stabilizing metaplastic phenotypes [231,232].
• Model Human Disease: Develop robust human-derived organoid models and improved in vivo systems that closely mimic the spectrum of gastric injury and metaplasia, to facilitate both mechanistic insight and therapeutic screening [233].

Understanding the mesenchymal niche is crucial, as it likely provides essential support and regulatory cues that influence gland development, epithelial cell differentiation, and responses to injury or disease. Further research is needed to elucidate the diverse cell types within the mesenchyme, their molecular signaling pathways, and how they contribute to maintaining gastric tissue homeostasis and driving pathological processes such as metaplasia or cancer progression. These findings need to be validated in future prospective studies, which would be useful in developing more objective and rational surveillance guidelines.

A deeper mechanistic comprehension of gastric epithelial plasticity will not only elucidate the origins of metaplasia but also identify novel therapeutic targets to prevent progression to dysplasia and carcinoma. Cellular reprogramming in the stomach should therefore be viewed not as a random consequence of injury, but as a highly regulated, dynamic process with both adaptive and pathological consequences [28,33,229].

Understanding the role of gastric epithelial cells in IM can aid in the early detection of biomarkers specific to metaplastic changes, enable targeted therapies through modulation of signaling pathways to prevent disease progression, and support the exploration of IM reversal. KLF4 is both a marker and functional driver of gastric IM. Its role in intestinal differentiation, coupled with its diagnostic value, makes it a promising target for early detection and therapeutic intervention in gastric precancerous conditions. OLFM4 is one of the leading candidate biomarkers for IM, as its expression is consistently upregulated in metaplastic glands and correlates with the presence of intestinalized epithelial cells [169].

Exploring OLFM4’s mechanistic involvement in gastric epithelial cell plasticity and metaplastic transformation may uncover novel therapeutic targets to halt or reverse the progression of IM toward GC. Advances in molecular imaging and noninvasive assays could leverage OLFM4 as a marker for early detection and risk stratification, ultimately improving patient outcomes through timely intervention.

Spatial transcriptomics and multiplex immunofluorescence together provide a comprehensive view of the metaplasia microenvironment by mapping gene expression and validating protein activity within intact tissue architecture. This integrated approach reveals spatial relationships between diverse cell types and signaling gradients, offering critical insights for developing biomarkers and targeted therapies [234].

Our awareness of gastric epithelial differentiation has evolved from simple hierarchical models to a more dynamic picture involving plasticity, lineage restriction, and region-specific stem cell strategies. Future studies using single-cell transcriptomics, live imaging, and improved lineage tracing systems will be critical for resolving outstanding questions and developing therapies for gastric diseases. By elucidating these interactions, we seek to identify potential biomarkers of metaplasia that could facilitate early detection, inform prevention strategies, and support the development of novel therapeutic approaches to advance future research and clinical practice [235].

7. Biases and Limitations in Metaplasia of the Gastric Mucosa Study

Conceptual ambiguity around terms like metaplasia, transdifferentiation, and paligenosis hinders comparison across studies, prompting a call for clearer definitions. Much of the mechanistic understanding comes from animal models, which require validation in human-relevant systems such as organoids and clinical samples. Lineage tracing reveals inconsistent evidence about the origins and reversibility of metaplasia, like SPEM, with implications for treatment strategies. Some candidate biomarkers, such as OLFM4, have shown promise in distinguishing IIM from CIM. The predictive value of OLFM4 for progression from IM to gastric cancer remains unvalidated in prospective follow-up studies. Additional challenges include limited biomarker specificity, underrepresentation of stromal and microbial components, technical model limitations, and a lack of diverse genomic datasets, all underscoring the need for integrative, personalized research approaches. Clinical and epidemiological heterogeneity, coupled with publication and population biases driven by prevalence and investigator concentration, underscores the need for diverse, standardized, and longitudinal datasets to strengthen reproducibility and global applicability.

8. Conclusions

Cellular plasticity involves the capacity of cells to define, modify, and re-establish their fate. Gastric epithelial cells demonstrate significant plasticity, enabling them to alter their differentiation states through processes like transdifferentiation, dedifferentiation, or impaired differentiation in response to various physiological and pathological stimuli. Gastric epithelial cells are central to the development of metaplasia through processes of plasticity, injury response, and altered differentiation [4]. Their flexibility results in diverse phenotypic manifestations, such as normal gastric mucosa, SPEM, and IM. Ongoing research into the cellular and molecular dynamics of SPEM and IM is essential for advancing GC prevention and treatment strategies. Metaplasia of the gastric mucosa is a prevalent precancerous lesion with significant implications for GC risk, especially in its SPEM and incomplete forms. Understanding the subtypes and associated risks is crucial for developing effective surveillance and management strategies to mitigate progression to GC. The clinical significance and utility of gastric epithelial cell plasticity still require further validation through comprehensive clinical trials and evaluations. In conclusion, future studies that deeply investigate gastricepithelial cell plasticity are essential to address unresolved questions and advance its translation into clinical practice.