Cancer stem cell plasticity drives therapy resistance in colorectal cancer

Abstract

Colorectal cancer (CRC) remains a leading cause of cancer-related mortality worldwide, primarily due to therapy resistance, tumor recurrence, and metastasis. Increasing evidence indicates that cancer stem cells (CSCs) are central drivers of tumor initiation, progression, and resistance to conventional therapies. CSCs exhibit key biological features, including self-renewal, differentiation capacity, and remarkable phenotypic plasticity, enabling dynamic adaptation under therapeutic pressure. Critical signaling pathways such as Wnt/β-catenin, Notch, and phosphatidylinositol 3-kinase/protein kinase B regulate CSC maintenance and survival. Recent advances have highlighted the concept of cellular plasticity, particularly the ability of non-stem cancer cells to reacquire stem-like properties, defined as revival CSCs. In addition, the tumor microenvironment plays a pivotal role in sustaining CSC function through stromal interactions, immune modulation, and extracellular matrix remodeling. Emerging therapeutic strategies targeting CSCs - including pathway inhibition, epigenetic modulation, and exosome-based therapies - offer promising approaches to overcome resistance. This mini-review summarizes current insights into CSC biology, resistance mechanisms, and evolving therapeutic strategies in CRC.

Key Words: Colorectal cancer; Cancer stem cells; Plasticity; Therapy resistance; Tumor microenvironment; Metastasis

Core Tip: Cancer stem cells (CSCs) are major drivers of therapy resistance, recurrence, and metastasis in colorectal cancer. Increasing evidence indicates that CSC plasticity enables differentiated tumor cells to reacquire stem-like properties, thereby sustaining tumor heterogeneity and therapeutic failure. This review highlights the molecular mechanisms regulating CSC plasticity, including developmental signaling pathways, epigenetic reprogramming, metabolic adaptation, and tumor microenvironment interactions. It also discusses emerging therapeutic strategies targeting CSCs and their plasticity, emphasizing that simultaneous inhibition of stemness and phenotypic switching may improve long-term treatment outcomes.

INTRODUCTION

Colorectal cancer (CRC) is one of the most frequently diagnosed malignancies worldwide and remains a leading cause of cancer-related mortality despite substantial advances in screening, surgical techniques, systemic chemotherapy, and targeted therapies. According to the latest GLOBOCAN 2022 estimates, CRC ranks among the most common cancers globally and continues to represent a major public health burden because of its high incidence and mortality rates, particularly in advanced-stage disease[1-4]. Treatment failure is frequently attributed to tumor recurrence, metastatic progression, and resistance to therapy. Despite improvements in multimodal treatment strategies, including combination chemotherapy regimens such as FOLFOX and FOLFIRI as well as biologic agents targeting epidermal growth factor receptor and vascular endothelial growth factor pathways, long-term survival remains limited in advanced-stage disease[5]. This highlights the persistence of therapy-resistant tumor cell populations that are not effectively eradicated by current treatment modalities. Increasing evidence suggests that these resistant populations are not static but dynamically regulated, contributing to disease relapse even after an initial therapeutic response[6].

Tumor heterogeneity is a hallmark of CRC and plays a crucial role in disease progression[7,8]. Within this heterogeneous tumor population, cancer stem cells (CSCs) represent a small but highly tumorigenic subpopulation capable of self-renewal and differentiation[9,10]. These cells are widely considered key drivers of tumor initiation, progression, and relapse[11,12]. Importantly, in epithelial malignancies, particularly CRC, accumulating evidence challenges the traditional hierarchical model of tumor organization and instead supports a dynamic model in which tumor cells can reversibly transition between stem-like and differentiated states. In contrast, the classical hierarchical CSC model remains well established in many hematologic malignancies. This phenotypic plasticity enables non-CSC populations to reacquire stemness properties under specific environmental or therapeutic pressures, thereby maintaining tumor heterogeneity and regenerative capacity[13]. In CRC, this dynamic equilibrium is closely linked to fluctuations in key signaling pathways such as Wnt/β-catenin, which are spatially and temporally regulated within the tumor microenvironment (TME)[14,15].

Importantly, CSCs exhibit intrinsic resistance to chemotherapy and radiotherapy through multiple mechanisms, including enhanced DNA repair capacity, resistance to apoptosis, and metabolic adaptation[16]. In addition to these intrinsic properties, CSCs can enter a quiescent or slow-cycling state, allowing them to evade cytotoxic agents that primarily target proliferating cells[17,18]. Furthermore, chromatin remodeling and epigenetic reprogramming have been shown to induce a reversible drug-tolerant state in cancer cell subpopulations, further contributing to therapeutic resistance[19,20]. Clinically, this contributes to minimal residual disease and metastatic recurrence, particularly in patients undergoing systemic treatment. Notably, colorectal CSCs have been shown to survive conventional chemotherapy and subsequently repopulate tumors, driving disease relapse and progression[21,22]. Therefore, targeting CSC biology, including their plasticity and interaction with the TME, has emerged as a critical strategy for improving long-term oncologic outcomes.

LITERATURE REVIEW

This narrative mini-review was prepared through a structured literature search focusing on recent advances in CRC stem cell biology, cellular plasticity, therapy resistance, and the TME. Electronic databases including PubMed, Scopus, and Web of Science were searched for English-language articles published up to May 2026. Search terms included combinations of “colorectal cancer”, “cancer stem cells”, “cellular plasticity”, “revival stem cells”, “tumor microenvironment”, “therapy resistance”, “Wnt signaling”, “Notch signaling”, “epigenetics”, and “exosomes”. Priority was given to high-quality original research articles, landmark studies, and recent review articles published in peer-reviewed journals. The literature was selected based on scientific relevance, methodological quality, novelty, and contribution to the understanding of CSC biology in CRC. As this work represents a narrative mini-review rather than a systematic review or meta-analysis, formal risk-of-bias assessment and quantitative data synthesis were not performed.

CSCS İN CRC

In CRC, CSCs are commonly identified by markers such as CD133, CD44, leucine-rich G repeat-containing protein-coupled receptor 5 (LGR5), and aldehyde dehydrogenase 1 (ALDH1)[21]. Among these, LGR5 has emerged as a particularly robust marker of intestinal stem cells and colorectal CSCs, supported by lineage-tracing studies demonstrating that LGR5+ cells are capable of sustaining long-term tumor growth and regeneration[23,24]. However, it is increasingly recognized that CSC marker expression is not static; rather, it is dynamically regulated, reflecting the plastic nature of tumor cell states. This dynamic marker expression complicates the identification of definitive CSC populations and suggests that stemness is a functional state rather than a fixed cellular identity. Experimental evidence supports this dynamic view. For example, lineage-tracing and organoid-based studies have shown that LGR5-negative CRC cells can reacquire LGR5 expression and regenerate tumor growth after depletion of LGR5-positive CSCs[24]. Similarly, studies on Wnt activity demonstrated that colorectal CSC states are spatially and functionally regulated by microenvironmental signals rather than being defined by fixed marker expression[14].

Beyond their tumor-initiating capacity, CSCs exhibit remarkable adaptability under hostile conditions such as hypoxia and therapeutic stress[22]. Hypoxic niches within the TME activate hypoxia-inducible factors (HIFs), which in turn promote stemness-associated transcriptional programs and metabolic reprogramming favoring glycolysis. In addition, CSCs display metabolic flexibility, enabling them to switch between oxidative phosphorylation (OXPHOS) and glycolysis depending on environmental conditions. This metabolic plasticity enhances survival under nutrient deprivation and contributes to resistance against chemotherapy and radiotherapy. Furthermore, interactions with stromal components, including cancer-associated fibroblasts (CAFs) and immune cells, reinforce CSC phenotypes through paracrine signaling mechanisms[25-27].

This adaptability is largely driven by cellular plasticity, allowing differentiated tumor cells to revert to a stem-like phenotype. This bidirectional interconversion significantly contributes to tumor recurrence and therapeutic resistance[13,28]. Mechanistically, this process is regulated by key developmental signaling pathways such as Wnt/β-catenin, Notch, and Hedgehog, which are frequently dysregulated in CRC[12,13]. Importantly, these pathways do not act in isolation but are modulated by the TME, enabling dynamic shifts between stem-like and differentiated states. Epigenetic mechanisms, including chromatin remodeling and DNA methylation changes, further facilitate this reversible reprogramming, allowing rapid adaptation to therapeutic stress[19,20]. This supports a non-hierarchical tumor model in which stemness can be reacquired rather than being restricted to a fixed subpopulation.

From a clinical perspective, this plasticity explains why elimination of bulk tumor cells often fails to achieve durable responses, as non-CSC populations can replenish the CSC pool following treatment. Indeed, therapeutic interventions may inadvertently promote CSC enrichment by inducing stress-response pathways and selecting for resistant phenotypes. This phenomenon is particularly evident in CRC patients receiving systemic chemotherapy, where residual disease often exhibits enhanced stem-like characteristics and increased tumorigenic potential[29,30]. Consequently, effective treatment strategies must not only target existing CSCs but also prevent the dedifferentiation of non-CSCs into stem-like states, highlighting the need for combination therapies that simultaneously address tumor bulk, CSC populations, and the underlying mechanisms of plasticity. A summary of the principal CRC stem cell markers, their biological functions, and their clinical significance is presented in Table 1. The biological characteristics and phenotypic plasticity of CRC stem cells are illustrated in Figure 1.

Figure 1 Biology of colorectal cancer stem cells. Normal intestinal epithelium contains leucine-rich G repeat-containing protein-coupled receptor 5+ intestinal stem cells that maintain epithelial homeostasis through self-renewal and differentiation. During colorectal tumorigenesis, a subset of tumor cells acquires cancer stem cell (CSC) characteristics, including self-renewal, differentiation capacity, and tumor-initiating potential. CSCs exhibit remarkable phenotypic plasticity, allowing bidirectional interconversion between CSCs and non-CSCs through dedifferentiation and reprogramming. Under chemotherapy, radiotherapy, or targeted therapy, resistant CSC populations are preferentially enriched, contributing to tumor recurrence, metastasis, and treatment failure. CSC: Cancer stem cell; LGR5: Leucine-rich G repeat-containing protein-coupled receptor 5.

Table 1 Major colorectal cancer stem cell markers and their biological significance.

Marker	Biological function	Clinical significance
CD133	Stem cell maintenance, tumor initiation	Associated with poor prognosis and chemoresistance
CD44	Cell adhesion, migration, CSC maintenance	Promotes invasion, metastasis, and therapy resistance
LGR5	Intestinal stem cell marker, Wnt signaling	Drives tumor initiation and regeneration
ALDH1	Detoxification and stem cell metabolism	Correlates with CSC enrichment and poor survival
EpCAM	Cell adhesion and proliferation	Frequently expressed in colorectal CSC populations

CSC: Cancer stem cell; LGR5: Leucine-rich G repeat-containing protein-coupled receptor 5; ALDH1: Aldehyde dehydrogenase 1; EpCAM: Epithelial cell adhesion molecule.

MOLECULAR MECHANİSMS OF CSC-MEDİATED RESİSTANCE

Key signaling pathways

Several signaling pathways play central roles in maintaining CSC properties, including Wnt/β-catenin, Notch, and phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT) signaling[14,15,30]. These pathways are not only essential for sustaining stemness but also dynamically regulated within the TME, enabling CSCs to adapt to therapeutic stress[6,31].

Wnt/β-catenin signaling is essential for stemness and tumor initiation, whereas Notch signaling regulates cell fate decisions and differentiation. The PI3K/AKT pathway promotes survival and resistance to apoptosis. In CRC, constitutive activation of Wnt signaling - frequently driven by APC mutations - creates a permissive environment for CSC maintenance[3,4], yet emerging evidence indicates that Wnt activity is heterogeneous and spatially regulated within tumors. This gradient of signaling activity allows for reversible transitions between high and low stemness states, reinforcing the concept of CSC plasticity[13,14]. Notch signaling further contributes to this dynamic by maintaining undifferentiated states and supporting resistance to differentiation-inducing therapies[30].

In addition, PI3K/AKT signaling integrates survival cues from growth factors and the TME, enhancing resistance to apoptosis and promoting metabolic adaptation under stress conditions. Crosstalk between these pathways amplifies CSC resilience; for example, PI3K/AKT activation can stabilize β-catenin signaling, while Notch signaling may synergize with Wnt to sustain stem-like phenotypes. Aberrant activation of these interconnected networks enhances CSC survival and contributes to resistance to conventional therapies[17,32]. Importantly, recent studies suggest that these pathways also regulate phenotypic switching, enabling non-CSCs to acquire stem-like characteristics following therapeutic pressure[6].

Despite their importance, therapeutic targeting of these pathways remains challenging due to pathway redundancy, compensatory feedback mechanisms, and systemic toxicity. Inhibiting a single pathway often leads to activation of alternative signaling routes, limiting therapeutic efficacy. Therefore, combinatorial strategies targeting multiple pathways or selectively modulating pathway activity within the CSC niche are being actively explored. Context-specific targeting approaches, guided by molecular profiling of tumors, may offer a more effective strategy to overcome resistance while minimizing adverse effects.

DNA damage response and repair

CSCs exhibit enhanced DNA repair capacity, enabling survival following genotoxic stress induced by chemotherapy and radiotherapy[19]. This enhanced DNA damage response (DDR) is mediated through upregulation of key repair pathways, including homologous recombination and non-homologous end joining, allowing CSCs to efficiently repair DNA double-strand breaks.

In addition to classical repair mechanisms, CSCs display increased activation of checkpoint kinases such as ATM and ATR, which facilitate cell cycle arrest and allow time for DNA repair. This coordinated response not only preserves genomic integrity but also prevents apoptosis, thereby promoting survival under therapeutic stress[6,19]. Furthermore, CSCs often maintain lower levels of reactive oxygen species (ROS) through enhanced antioxidant defenses, reducing DNA damage accumulation and contributing to radioresistance.

This feature has important clinical implications, particularly in the context of neoadjuvant therapy and radiotherapy resistance in rectal cancer. Patients with tumors enriched in CSC populations frequently demonstrate incomplete response to chemoradiotherapy, leading to residual disease and increased risk of local recurrence[30]. These observations underscore the need for therapeutic strategies targeting DDR pathways in CSCs, such as inhibitors of checkpoint kinases or DNA repair enzymes, which may sensitize these cells to conventional treatments.

Epigenetic regulation

Epigenetic mechanisms play a crucial role in regulating CSC phenotypes[18]. Histone modifications, DNA methylation patterns, and chromatin remodeling collectively influence gene expression programs associated with stemness, plasticity, and resistance[20]. Unlike genetic mutations, these epigenetic changes are reversible, allowing CSCs to rapidly adapt to environmental and therapeutic pressures.

Epigenetic regulators such as SMYD3 have been shown to activate oncogenic pathways, including c-MYC, thereby promoting CSC maintenance and tumor progression[33]. In addition, chromatin remodeling complexes can induce a drug-tolerant persister state, characterized by transient resistance to therapy without permanent genetic alterations[19,20]. This reversible state is particularly important in explaining minimal residual disease and tumor relapse following initial treatment response.

Moreover, epigenetic plasticity is closely linked to transcriptional programs associated with epithelial-mesenchymal transition (EMT), further reinforcing stem-like phenotypes and invasive potential[34-36]. DNA methylation changes and histone modifications can activate EMT-related transcription factors, facilitating phenotypic switching and enhancing metastatic capacity. From a therapeutic perspective, targeting epigenetic regulators using histone deacetylase (HDAC) inhibitors or DNA methyltransferase (DNMT) inhibitors represents a promising approach to disrupt CSC plasticity and sensitize tumors to standard treatments. However, clinical translation remains challenging due to off-target effects and the complexity of epigenetic networks.

METABOLİC REPROGRAMMİNG AND THERAPY RESİSTANCE

Metabolic reprogramming is increasingly recognized as a fundamental hallmark of CSCs and represents a key determinant of CRC initiation, progression, metastasis, and resistance to therapy. Unlike differentiated tumor cells, CSCs exhibit remarkable metabolic plasticity, allowing them to dynamically shift between glycolysis and mitochondrial OXPHOS depending on oxygen availability, nutrient status, and therapeutic stress. This metabolic flexibility enables CSCs to maintain energy production, preserve redox balance, and survive under hostile microenvironmental conditions, thereby promoting long-term tumor persistence and recurrence[37-39].

During tumor initiation, oncogenic signaling pathways including Wnt/β-catenin, PI3K/AKT/mechanistic target of rapamycin (mTOR), MYC, and HIF-1α cooperate to reprogram cellular metabolism toward anabolic growth and stemness maintenance. Activation of HIF-1α under hypoxic conditions promotes glycolytic enzyme expression while suppressing mitochondrial respiration, enabling CSC survival within poorly vascularized tumor niches. Conversely, subsets of colorectal CSCs preferentially rely on OXPHOS, demonstrating that metabolic heterogeneity is an essential feature of CSC biology rather than a fixed metabolic phenotype[38-41].

As colorectal tumors evolve, CSCs further exploit alternative metabolic pathways including glutamine metabolism, fatty acid β-oxidation, and lipid biosynthesis to support proliferation, EMT, metastatic dissemination, and immune evasion. Enhanced mitochondrial biogenesis together with increased antioxidant capacity allows CSCs to efficiently detoxify ROS, thereby protecting genomic integrity during chemotherapy and radiotherapy[39-42].

Importantly, mounting evidence suggests that metabolic adaptation directly contributes to therapeutic resistance. Following exposure to cytotoxic agents, surviving CSCs frequently undergo metabolic switching toward mitochondrial respiration, increasing ATP production and activating stress-response pathways that promote drug tolerance. This metabolic rewiring also facilitates the emergence of therapy-induced stem-like cells, thereby contributing to tumor recurrence after apparently successful treatment[40-43].

Consequently, targeting metabolic vulnerabilities has emerged as a promising therapeutic strategy. Pharmacological inhibition of mitochondrial respiration, glutamine utilization, fatty acid oxidation, or key metabolic regulators such as AMP-activated protein kinase and mTOR may sensitize CSCs to chemotherapy and reduce tumor relapse. Combining metabolism-targeted therapies with conventional cytotoxic agents or immunotherapy may therefore represent an effective strategy to overcome CSC-mediated resistance in CRC[38,41-44].

TUMOR MİCROENVİRONMENT AND CSC INTERACTİON

The TME is a critical regulator of CSC function and behavior[25]. Beyond serving as a passive structural framework, the TME actively shapes CSC phenotypes through biochemical and mechanical cues[31,45]. Components such as CAFs, immune cells, endothelial cells, and extracellular matrix (ECM) interact with CSCs to promote tumor growth, invasion, and metastasis[26,31,45]. In CRC, these interactions are highly dynamic and spatially organized, creating specialized “CSC niches” that sustain stemness and protect tumor cells from therapeutic stress[25,27]. Importantly, these niches are not fixed; rather, they are continuously remodeled in response to inflammation, hypoxia, and treatment-induced changes[6,31].

CAFs represent one of the most influential stromal components regulating CSC behavior[27,32]. Through the secretion of cytokines and growth factors such as transforming growth factor (TGF)-β, interleukin (IL)-6, and hepatocyte growth factor, CAFs activate key signaling pathways including signal transducer and activator of transcription-3 (STAT3), Wnt/β-catenin, and PI3K/AKT, thereby reinforcing CSC self-renewal and survival[27,32]. Additionally, CAF-derived ECM remodeling increases tissue stiffness and alters mechanotransduction pathways, further enhancing stem-like phenotypes[31]. These mechanical signals, mediated through integrins and focal adhesion kinases, contribute to invasion and metastatic potential[31,46].

Stromal signaling pathways, including fibronectin (FN1)-CD44 interactions, facilitate CSC-mediated invasion and metastatic dissemination. The interaction between FN1 and CD44 not only enhances adhesion and migration but also activates downstream signaling cascades that promote survival and resistance[14,26]. Furthermore, hypoxic conditions within the TME play a central role in maintaining CSC populations by activating HIFs, which drive transcriptional programs associated with stemness, angiogenesis, and metabolic adaptation[6,31]. Hypoxia-induced signaling also promotes EMT, linking microenvironmental stress to increased plasticity and metastatic potential[34-36].

Immune modulation within the TME enables CSCs to evade immune surveillance, contributing to disease progression and resistance to immunotherapy[37]. CSCs can downregulate major histocompatibility complex expression and upregulate immune checkpoint molecules such as programmed death-ligand 1, thereby suppressing cytotoxic T-cell activity[37]. In addition, CSCs actively recruit immunosuppressive cell populations, including regulatory T cells and myeloid-derived suppressor cells (MDSCs), creating an immune-privileged niche[37]. Tumor-associated macrophages (TAMs), particularly the M2-polarized subtype, further support CSC maintenance through secretion of anti-inflammatory cytokines and growth factors[37]. This complex immunosuppressive network represents a major barrier to effective immunotherapy in CRC, particularly in microsatellite stable (MSS) tumors where immune checkpoint blockade has limited efficacy[47-49].

Targeting CSC-TME interactions is increasingly recognized as a promising approach to inhibit tumor progression[27,32]. Therapeutic strategies aimed at disrupting stromal signaling, remodeling the ECM, or reprogramming the immune microenvironment are under active investigation[27,32]. Combination approaches integrating CSC-targeted therapies with immunotherapy or anti-stromal agents may enhance treatment efficacy by simultaneously eliminating CSCs and dismantling their supportive niche[47-49]. However, the heterogeneity and adaptability of the TME remain significant challenges, underscoring the need for personalized and context-specific therapeutic strategies[6].

Chronic inflammation is increasingly recognized as a fundamental driver of colorectal tumorigenesis and CSC maintenance. Persistent inflammatory conditions promote a protumorigenic microenvironment through the continuous production of cytokines such as tumor necrosis factor-α, IL-1β, and IL-6, which activate nuclear factor kappa B (NF-κB) signaling. NF-κB functions as a master transcriptional regulator linking inflammation to cancer progression by inducing the expression of genes involved in cell survival, proliferation, angiogenesis, EMT, and immune evasion. Importantly, constitutive NF-κB activation has been associated with the maintenance of CSC phenotypes, enhancement of cellular plasticity, and increased resistance to chemotherapy and radiotherapy through extensive crosstalk with STAT3, Wnt/β-catenin, and PI3K/AKT signaling pathways. Furthermore, inflammatory signaling within the TME promotes dedifferentiation of non-CSCs into stem-like cells, thereby facilitating tumor recurrence following treatment. Consequently, therapeutic strategies targeting inflammatory mediators or NF-κB signaling may disrupt CSC maintenance and improve the efficacy of conventional anticancer therapies[50-54].

CELLULAR PLASTİCİTY AND REVİVAL CSCS

The recognition of cellular plasticity represents one of the most significant advances in cancer biology[28]. CSCs are not a static population; instead, tumor cells dynamically transition between stem-like and differentiated states[6,7]. This dynamic equilibrium challenges the traditional hierarchical model of tumor organization and supports a more flexible, bidirectional model in which cellular identity is continuously reshaped by intrinsic signaling pathways and extrinsic microenvironmental cues[6]. In CRC, this plasticity is particularly pronounced due to the inherent regenerative capacity of intestinal epithelium, where normal stem cell programs can be hijacked by tumor cells to sustain growth and adaptation[15,23].

At the molecular level, cellular plasticity is governed by the coordinated activity of developmental pathways such as Wnt/β-catenin, Notch, and TGF-β signaling, along with epigenetic regulators that enable rapid transcriptional reprogramming[14,31,32]. Fluctuations in Wnt signaling activity, for instance, have been shown to define distinct stemness states within tumors, allowing cells to reversibly enter or exit a stem-like phenotype[3,14]. In addition, chromatin remodeling and histone modification patterns facilitate transitions into drug-tolerant states, further enhancing the adaptability of tumor cells under therapeutic pressure[19,20]. These reversible changes are central to the persistence of minimal residual disease[6].

The concept of revival CSCs (revCSCs) describes the ability of differentiated tumor cells to reacquire stemness following therapeutic intervention[13,28]. This process is not merely a passive survival mechanism but an active, regulated response to environmental stress, including chemotherapy, radiotherapy, and hypoxia[6]. Recent evidence suggests that injury-like signaling pathways, similar to those activated during tissue regeneration, play a crucial role in this process[13,23]. In particular, inflammatory signals and niche-derived factors can induce a “revival” program in differentiated cells, reactivating stem cell-associated gene expression and restoring tumor-initiating capacity[25,27]. This highlights the close parallel between tumor regeneration and normal tissue repair mechanisms[23].

Importantly, EMT has been closely linked to cellular plasticity and the emergence of revCSCs[22-24]. During EMT, tumor cells acquire enhanced migratory capacity and stem-like properties, enabling both dissemination and survival under therapeutic stress[34-36]. However, rather than representing a binary switch, EMT exists along a spectrum of intermediate states, which are increasingly recognized as highly plastic and functionally relevant[34,35]. These hybrid phenotypes exhibit both epithelial and mesenchymal features and are associated with maximal tumorigenic potential and resistance[34,35].

Clinically, this plasticity explains why elimination of bulk tumor cells or even pre-existing CSC populations often fails to achieve durable responses, as non-CSC populations can replenish the CSC pool following treatment[13]. This phenomenon is particularly relevant in CRC patients receiving systemic chemotherapy, where initial tumor shrinkage is frequently followed by relapse driven by reprogrammed, stem-like cells[29,30]. Moreover, therapeutic interventions themselves may induce plasticity by activating stress-response pathways, thereby inadvertently promoting CSC regeneration[6].

Therefore, this highlights the need for treatment strategies that inhibit both CSCs and plasticity mechanisms. Targeting the regulatory networks that enable phenotypic switching - such as epigenetic modulators, EMT-associated pathways, and niche-derived signaling - may be essential to prevent tumor relapse[20]. Combination therapies designed to simultaneously eliminate CSCs and block dedifferentiation processes represent a promising approach for improving long-term outcomes in CRC[6]. The molecular mechanisms regulating CSC plasticity and therapy resistance are summarized in Figure 2.

Figure 2 Molecular mechanisms driving cancer stem cell plasticity and therapy resistance in colorectal cancer. Multiple developmental signaling pathways, including Wnt/β-catenin, Notch, Hedgehog, phosphatidylinositol 3-kinase/protein kinase B/mechanistic target of rapamycin, and transforming growth factor-β signaling, cooperatively regulate cancer stem cell (CSC) maintenance, self-renewal, and phenotypic plasticity. The tumor microenvironment, composed of cancer-associated fibroblasts, immune cells, endothelial cells, cytokines, and extracellular matrix, further supports CSC survival and stemness. Additional mechanisms, including enhanced DNA damage repair, anti-apoptotic signaling, metabolic reprogramming, and epigenetic regulation, promote resistance to chemotherapy and radiotherapy, leading to minimal residual disease, tumor recurrence, and metastasis. CSC: Cancer stem cell; PI3K: Phosphatidylinositol 3-kinase; AKT: Protein kinase B; mTOR: Mechanistic target of rapamycin; TGF: Transforming growth factor; EMT: Epithelial-mesenchymal transition; CAF: Cancer-associated fibroblast; ECM: Extracellular matrix; OXPHOS: Oxidative phosphorylation.

EMERGING THERAPEUTIC STRATEGIES

Targeting signaling pathways

Inhibitors targeting Wnt, Notch, and PI3K/AKT pathways aim to disrupt CSC maintenance and survival[15,30]. These pathways are central regulators of stemness and are frequently dysregulated in CRC, making them attractive therapeutic targets[3,14]. However, clinical translation has been limited due to pathway redundancy and compensatory signaling mechanisms[6,30].

For instance, inhibition of Wnt signaling - while theoretically effective in reducing CSC populations - has been associated with significant toxicity due to its critical role in normal intestinal stem cell homeostasis[15,23]. Similarly, Notch pathway inhibitors, including γ-secretase inhibitors, have demonstrated limited clinical success due to gastrointestinal toxicity and incomplete pathway suppression[13]. The PI3K/AKT pathway, although widely targeted, often exhibits feedback activation of parallel survival pathways, diminishing therapeutic efficacy[6].

Importantly, recent studies suggest that single-agent targeting of these pathways is insufficient due to extensive crosstalk and adaptive resistance mechanisms[5]. As a result, combination strategies targeting multiple signaling axes simultaneously, or integrating pathway inhibitors with chemotherapy or immunotherapy, are increasingly being explored[47-49]. Precision medicine approaches, guided by tumor-specific molecular profiling, may further enhance the efficacy of these strategies by identifying patients most likely to benefit from pathway-targeted interventions[55].

Epigenetic therapy

Epigenetic therapies have the potential to reverse CSC phenotypes and sensitize tumors to conventional treatments[18]. Given the reversible nature of epigenetic modifications, targeting chromatin regulators represents a promising strategy to disrupt CSC plasticity and eliminate drug-tolerant cell populations[20].

HDAC inhibitors and DNMT inhibitors have been shown to alter gene expression profiles associated with stemness, differentiation, and apoptosis[20]. These agents can promote differentiation of CSCs into more therapy-sensitive states, thereby enhancing the efficacy of cytotoxic treatments[18]. In addition, epigenetic therapies may interfere with the formation of drug-tolerant persister cells by preventing adaptive transcriptional reprogramming[19,20].

Furthermore, emerging evidence indicates that epigenetic modulation can enhance anti-tumor immunity by increasing tumor antigen presentation and reducing immune evasion[37]. This has led to growing interest in combining epigenetic therapies with immune checkpoint inhibitors, particularly in tumors that are otherwise resistant to immunotherapy[47-49,56]. However, challenges remain in optimizing dosing strategies and minimizing off-target effects, as epigenetic regulators often influence a broad range of cellular processes.

TME modulation

Targeting CSC-TME interactions may reduce tumor progression, metastasis, and therapy resistance[27]. Given the critical role of the microenvironment in sustaining CSC phenotypes, therapeutic strategies aimed at disrupting stromal support systems are gaining increasing attention[31].

Approaches targeting CAFs, ECM remodeling, and key cytokine signaling pathways - such as IL-6/STAT3 and TGF-β - have demonstrated potential in preclinical models[27,32]. Inhibition of these pathways can reduce CSC self-renewal capacity and impair metastatic dissemination[27]. Additionally, normalization of the TME, including modulation of hypoxia and angiogenesis, may further enhance treatment response[31].

Immune-based strategies also play a central role in TME modulation. Although immune checkpoint inhibitors have shown limited efficacy in MSS CRC, combining these agents with therapies targeting CSCs or stromal components may overcome resistance[39-41]. Reprogramming immunosuppressive cell populations such as TAMs and MDSCs represents another promising avenue to restore anti-tumor immunity and disrupt CSC niches[45].

Exosome-based therapies

Mesenchymal stem cell-derived exosomes (MSC-Exo) have demonstrated immunomodulatory and anti-tumor effects[1]. These extracellular vesicles serve as important mediators of intercellular communication, carrying proteins, lipids, and nucleic acids that can influence tumor behavior and immune responses[1,37].

MSC-Exos have been shown to modulate inflammation, inhibit tumor proliferation, and alter the TME in ways that may suppress CSC activity[1,27]. In addition, engineered exosomes are being explored as delivery vehicles for therapeutic agents, including small interfering RNAs, microRNAs, and chemotherapeutic drugs[1]. This targeted delivery approach offers the potential to selectively modulate CSC-related pathways while minimizing systemic toxicity.

However, the role of exosomes in cancer is complex and context-dependent, as they may also promote tumor progression under certain conditions[1,45]. Therefore, a deeper understanding of exosome biology and standardization of production and delivery methods are essential before widespread clinical application. Despite these challenges, exosome-based therapies represent a rapidly evolving and promising frontier in the treatment of CRC. A schematic overview of the current and emerging therapeutic strategies targeting CRC stem cells is presented in Figure 3.

Figure 3 Therapeutic strategies targeting colorectal cancer stem cells. Conventional therapies primarily eliminate rapidly proliferating non-cancer stem cells (CSCs) but frequently fail to eradicate therapy-resistant CSCs, resulting in tumor relapse and metastatic progression. Emerging combination strategies aim to overcome CSC-mediated resistance by simultaneously targeting CSC-associated signaling pathways, epigenetic regulators, tumor microenvironment components, immune checkpoints, and exosome-mediated communication. These multimodal therapeutic approaches are expected to inhibit CSC plasticity, prevent dedifferentiation, improve treatment response, and achieve durable clinical outcomes. In the final panel, eliminated/apoptotic CSCs represent treatment-induced cancer stem cell death, whereas the remaining differentiated cells represent tumor cells that have lost stem-like properties and entered a more treatment-sensitive state. The reduction in CSCs and residual tumor cells illustrates the intended durable therapeutic response. Symbols and colors are defined within the figure. CSC: Cancer stem cell; PI3K: Phosphatidylinositol 3-kinase; AKT: Protein kinase B; mTOR: Mechanistic target of rapamycin; HDAC: Histone deacetylase; CAF: Cancer-associated fibroblast; ECM: Extracellular matrix; CAR-T: Chimeric antigen receptor T; siRNA: Small interfering RNA; miRNA: MicroRNA.

FUTURE PERSPECTİVES

Future research should focus on several critical areas that address the dynamic and adaptive nature of CSCs in CRC. Targeting CSC plasticity mechanisms remains a central priority. Rather than viewing CSCs as a fixed subpopulation, future strategies must consider stemness as a reversible and context-dependent state. Elucidating the molecular regulators of phenotypic switching - particularly those governing transitions between differentiated tumor cells and stem-like states - will be essential. This includes deeper investigation into epigenetic reprogramming, EMT-associated transcriptional networks, and microenvironment-driven signaling pathways. Therapeutic interventions designed to inhibit these dynamic transitions may prove more effective than approaches targeting static CSC markers.

Combining CSC-targeted therapies with conventional treatments is likely to represent a more effective therapeutic paradigm. Monotherapies directed against CSC-associated pathways have shown limited success due to redundancy and compensatory mechanisms. Therefore, integrating CSC-targeted agents with chemotherapy, radiotherapy, and immunotherapy may enhance treatment response by simultaneously eliminating tumor bulk and preventing CSC-mediated relapse. Rational design of combination therapies, guided by tumor biology and treatment timing, will be critical to maximize efficacy while minimizing toxicity.

Identifying reliable CSC biomarkers is another key challenge. Current markers such as CD133, CD44, and LGR5 lack specificity and may not fully capture the dynamic nature of CSC states. Advances in single-cell RNA sequencing and spatial transcriptomics offer promising tools for identifying functional CSC populations and tracking their evolution during treatment. The development of robust, clinically applicable biomarkers will enable better patient stratification, real-time monitoring of therapeutic response, and early detection of resistance.

Developing personalized therapeutic strategies is essential in the context of tumor heterogeneity and plasticity. Inter-patient and intra-tumoral variability necessitates individualized treatment approaches based on molecular profiling and dynamic tumor behavior. Integrating multi-omics data, including genomic, transcriptomic, and epigenomic information, may allow for more precise targeting of CSC-related pathways and improved prediction of treatment response.

Beyond these priorities, future studies should also explore the integration of emerging technologies such as lineage tracing, organoid models, and artificial intelligence-driven predictive modeling to better understand CSC dynamics. These approaches may provide unprecedented insights into tumor evolution, treatment adaptation, and mechanisms of resistance.

Ultimately, understanding CSC dynamics - including their interaction with the TME, their capacity for phenotypic plasticity, and their role in therapy resistance - will be critical for improving clinical outcomes in CRC. Translating these insights into effective therapeutic strategies represents one of the most important challenges and opportunities in modern oncology.

In addition to identifying functional CSC populations, future research should focus on biomarkers capable of capturing dynamic cellular transitions, including dedifferentiating and highly plastic tumor cell states. Emerging approaches such as single-cell transcriptomics, spatial transcriptomics, lineage-tracing technologies, and multi-omics profiling may facilitate the identification of transient cellular states associated with CSC reprogramming and therapy-induced plasticity. Such biomarkers could improve early detection of treatment resistance and enable more precise patient stratification for CSC-targeted therapeutic interventions[13,28]. The major therapeutic approaches currently being investigated for targeting CRC stem cells are summarized in Table 2.

Table 2 Emerging therapeutic strategies targeting colorectal cancer stem cells.

Therapeutic strategy	Primary target	Proposed mechanism	Current status
Wnt inhibitors	Wnt/β-catenin	Reduce CSC self-renewal	Preclinical/early clinical
Notch inhibitors	Notch signaling	Promote CSC differentiation	Clinical evaluation
PI3K/AKT inhibitors	PI3K/AKT pathway	Inhibit survival signaling	Clinical evaluation
HDAC inhibitors	Histone acetylation	Reverse stemness-associated epigenetic changes	Clinical trials
DNMT inhibitors	DNA methylation	Restore differentiation programs	Clinical trials
Immune checkpoint inhibitors	PD-1/PD-L1	Enhance anti-tumor immunity	Approved for MSI-H CRC
Exosome-based therapies	CSC communication	Drug delivery and microenvironment modulation	Experimental

CSC: Cancer stem cell; PI3K: Phosphoinositide 3-kinases; AKT: Protein kinase B; HDAC: Histone deacetylase; DNMT: DNA methyltransferase; PD-1: Programmed cell death protein 1; PD-L1: Programmed death-ligand 1; MSI-H: Microsatellite instability-high; CRC: Colorectal cancer.

Limitations of this review

This review has several limitations that should be acknowledged. First, as a narrative mini-review, it does not follow a systematic review methodology or include formal quality assessment or risk-of-bias analysis of the included studies. Second, the discussion is primarily based on published preclinical and translational studies, while clinical evidence regarding CSC-targeted therapies remains limited and continues to evolve. Third, although major signaling pathways, cellular plasticity, TME interactions, metabolic reprogramming, and emerging therapeutic strategies are discussed, the rapidly expanding nature of this field means that some recently identified molecular mechanisms and therapeutic approaches may not have been comprehensively covered. Finally, because CRC exhibits substantial molecular and clinical heterogeneity, the biological behavior of CSCs and treatment responses may vary across different patient populations and molecular subtypes. Therefore, continued clinical validation and prospective studies are required to translate these findings into routine clinical practice.

CONCLUSION

CSCs play a central role in CRC progression, therapy resistance, and recurrence. Beyond their well-established capacity for self-renewal and tumor initiation, CSCs represent a highly dynamic and adaptive cell population capable of responding to therapeutic and microenvironmental pressures. This adaptability is largely driven by cellular plasticity, which enables bidirectional transitions between stem-like and differentiated states, thereby sustaining tumor heterogeneity and facilitating disease relapse.

Advances in understanding CSC biology, particularly cellular plasticity and TME interactions, have revealed novel therapeutic opportunities. The recognition that CSCs are not a fixed entity but rather a functional and reversible state has fundamentally reshaped current paradigms of tumor organization and treatment resistance. Interactions between CSCs and their niche - including stromal cells, immune components, and ECM - further enhance their survival, metastatic potential, and resistance to conventional and targeted therapies.

Importantly, therapeutic strategies that focus solely on eliminating bulk tumor cells are unlikely to achieve durable responses, as residual or dedifferentiated cells can regenerate the CSC pool. Therefore, effective treatment approaches must simultaneously target CSC populations, inhibit plasticity-driven phenotypic switching, and disrupt the supportive TME. Emerging combination strategies integrating CSC-targeted agents with chemotherapy, immunotherapy, and microenvironment modulation hold significant promise in this regard.

Ultimately, a comprehensive understanding of CSC dynamics - including their molecular regulation, interaction with the TME, and role in therapy resistance - will be essential for the development of more effective and durable treatment strategies. Translating these insights into clinical practice has the potential to significantly improve patient survival and represents a critical frontier in CRC research.

ACKNOWLEDGEMENTS

I would also like to thank Mustafa Duman, Erdogan Mutevelli Sozuer, Hızır Yakup Akyıldız, Hasan Bektas, Gurhan Sakman, and Cem Kaan Parsak for enabling me to complete my training in molecular oncology, surgical oncology, and general surgery.