Natural killer cell-based immunotherapies for colorectal cancer: Current strategies, challenges, and future perspectives

Abstract

Natural killer (NK) cells have emerged as promising therapeutic agents for treating colorectal cancer because of their innate ability to recognize and eliminate tumor cells without prior sensitization. In this review, NK cell-based immunotherapeutic approaches, including cytokine therapy, immune checkpoint inhibition, antibody-dependent cellular cytotoxicity, and adoptive cell transfer, are comprehensively examined, and their respective clinical potential and limitations are highlighted. We discuss critical challenges in NK cell expansion, genetic engineering (particularly chimeric antigen receptor-NK development), and tumor microenvironment-mediated immunosuppression. Furthermore, we explore innovative strategies such as combination therapies, nanotechnology-enhanced delivery, and personalized medicine approaches that aim to overcome the current barriers. The review concludes with future directions emphasizing the need for standardized manufacturing protocols, new strategies to improve NK cell persistence, and clinical validation of emerging technologies, positioning NK cell immunotherapy as a transformative modality for colorectal cancer treatment.

Key Words: Natural killer cells; Colorectal cancer; Immunotherapy; Immunosuppression; Tumor microenvironment

Core Tip: Natural killer (NK) cell-based immunotherapy represents a breakthrough approach for treating colorectal cancer (CRC), particularly for microsatellite-stable tumors resistant to conventional treatments. Currently employed strategies, including cytokine therapy, checkpoint inhibition, and chimeric antigen receptor-NK cell therapy, show promise in CRC treatment, but manufacturing challenges and the immunosuppressive tumor microenvironment limit their efficacy. Emerging solutions combining gene-editing, nanotechnology, and personalized approaches may overcome these barriers, positioning NK cell therapy as a transformative CRC treatment modality.

INTRODUCTION

Colorectal cancer (CRC), a type of cancer that develops in the colon or rectum and involves innate immunity[1], remains a formidable global health burden and ranks as the third most prevalent malignancy and the second leading cause of cancer-related death[2]. In 2022, over 1.9 million new CRC cases and 0.9 million CRC-related deaths were reported, accounting for approximately one-tenth of all cancer cases and cancer-related deaths[3,4]. While CRC incidence rates have historically been the highest in developed nations due to obesity-promoting diets and sedentary lifestyles, recent epidemiological reports have revealed an alarming rise in CRC incidence in Asia and Africa, which is a consequence of rapid urbanization and insufficient early screening infrastructure[2,3]. Extensive clinical evidence indicates that systematic screening initiatives combined with comprehensive preventive measures lead to a marked reduction in CRC incidence and associated mortality, which emphasizes the vital importance of early diagnosis and timely therapeutic interventions[5,6]. Notably, many CRC-related deaths occur in low- and middle-income countries, which underscores disparities in preventive care accessibility[2]. While established screening modalities (e.g., colonoscopy and fecal immunochemical testing) have demonstrated efficacy in reducing CRC incidence and mortality[7-9], significant therapeutic challenges remain. Notably, approximately 95% of CRC cases are microsatellite-stable tumors, which exhibit inherent resistance to conventional immune checkpoint inhibitors that target programmed cell death protein 1/programmed death ligand 1 (PD-1/PD-L1)[10]. Furthermore, acquired resistance frequently leads to disease progression even in initial responders. These clinical realities underscore the critical need to develop novel immunotherapeutic approaches for CRC treatment.

Natural killer (NK) cells, derived from CD34+ hematopoietic stem cells in the bone marrow, play a pivotal role in the innate immune system. Their development is tightly regulated by cytokine networks and transcriptional checkpoints[11]. For example, interleukin-15 (IL-15) plays a crucial role in promoting the expansion of NK cell precursors through activation of the signal transducer and activator of transcription 5 signaling pathway; concurrently, the helix-loop-helix protein ID2 protein regulates NK cell development by adjusting their responsiveness to IL-15[12]. After maturation, NK cells diverge into two functionally distinct subsets: (1) Circulating CD16+CD56^dim cells, which mediate antibody-dependent cellular cytotoxicity (ADCC); and (2) Long-term tissue-resident CD49a+CD103+ NK cells, which constitutively express tumor necrosis factor (TNF)-related apoptosis-inducing ligand and interferon-gamma (IFN-γ) for localized immunosurveillance[13,14]. Unlike adaptive lymphocytes, which express somatically rearranged antigen receptors, NK cells execute rapid effector functions without prior antigen sensitization[15], which are governed by a dynamic equilibrium between activating receptors [e.g., NK group 2, member D (NKG2D) and DNAX accessory molecule-1 (DNAM-1)] and inhibitory receptors (e.g., killer cell immunoglobulin like receptor and NKG2A) (Figure 1)[16,17]. Specifically, activating receptors initiate immune responses by recruiting kinases such as spleen tyrosine kinase or zeta-chain associated protein 70 through immunoreceptor tyrosine-based activation motifs, thereby triggering cellular activation; conversely, inhibitory receptors maintain immune homeostasis by recruiting phosphatases including Src homology domain 2-containing tyrosine phosphatase-1, via immunoreceptor tyrosine-based inhibitory motifs, which dephosphorylate key signaling molecules to suppress activation pathways[18]. Target cell recognition triggers a triphasic cytotoxic cascade, including perforin-mediated membrane pore formation, granzyme B-induced caspase-3 activation via mitochondrial permeability transition[19], and IFN-γ/TNF-α secretion to orchestrate other immune cell functions, such as dendritic cell maturation and T helper type 1 (Th1) polarization[20,21]. Low-affinity immunoglobulin G Fc region receptor III (CD16) plays a crucial role in NK cell-mediated immunity by binding to the Fc region of immunoglobulin G antibodies on opsonized target cells, inducing receptor cross-linking that drives ADCC and facilitates efficient target cell elimination[22,23]. For example, CD16-dependent ADCC enhances tumor targeting precision when coupled with monoclonal antibodies, positioning NK cells as versatile effectors against CRC[24]. A comprehensive summary of these regulatory mechanisms, including the major activating and inhibitory receptors, their cognate ligands, and their molecular interactions, is presented in Table 1[22,25-35]. Emerging evidence suggests that NK cell dysfunction is involved in CRC progression since the number of tumor-infiltrating NK cells may be correlated with metastatic potential[36-38]. This paradigm highlights the therapeutic potential of reinvigorating NK cell activity, a strategy now being harnessed through adoptive transfer, cytokine priming, and genetic engineering.

Figure 1 Receptor-mediated interactions between natural killer cells and colorectal cancer cells. Natural killer (NK) cells engage with colorectal cancer (CRC) cells through a complex network of activating and inhibitory receptor-ligand interactions[16,17,22,25-36]. The activating receptors on NK cells, including CD16, NK group 2, member D, NKp46, NKp30, NKp44, and DNAX accessory molecule-1, recognize specific ligands expressed on CRC cells. These ligands include CD112, CD155, tumor-associated glycoproteins, human leukocyte antigen-B-associated transcript 3, BT-H6, major histocompatibility complex class I chain-related proteins A and B, UL16-binding proteins 1-6, and tumor-associated antigens. Upon receptor-ligand binding, NK cells become activated and initiate antitumor responses through multiple mechanisms: (1) The upregulation of Fas ligand; (2) The release of cytotoxic granules containing perforin and granzyme B; and (3) The secretion of pro-inflammatory cytokines such as interferon-γ and tumor necrosis factor-α. These effector functions collectively lead to CRC cell death. Conversely, inhibitory receptors on NK cells, including killer-cell immunoglobulin-like receptors, NK group 2, member A, and tyrosine-based inhibitory motif domain, interact with their corresponding ligands (major histocompatibility complex class I molecules, human leukocyte antigen-E, and CD155, respectively) on CRC cells. These interactions drive inhibitory signals that suppress NK cell activation and consequently impair their antitumor cytotoxicity. NK: Natural killer; CRC: Colorectal cancer; TAA: Tumor-associated antigen; MICA/B: Major histocompatibility complex class I chain-related proteins A and B; ULBP1-6: UL16-binding proteins 1-6; TAG: Tumor-associated glycoprotein; BAT3: Human leukocyte antigen-B-associated transcript 3; MHC-I: Major histocompatibility complex class I; HLA: Human leukocyte antigen; NKG2D: Natural killer group 2, member D; DAP10: DNAX-activating protein of 10 kDa; FCR: Fc receptor; DAP12: DNAX-activating protein of 12 kDa; DNAM-1: DNAX accessory molecule-1; KIRs: Killer-cell immunoglobulin-like receptors; TIGIT: Tyrosine-based inhibitory motif domain; FasL: Fas ligand; GZMB: Granules containing perforin and granzyme B; IFN-γ: Interferon-gamma; TNF-α: Tumor necrosis factor-α; Syk: Spleen tyrosine kinase; ZAP70: Zeta-chain associated protein 70; SHP: Src homology 2 domain-containing phosphatase.

Table 1 Regulatory mechanisms of natural killer cell function: Key activating and inhibitory receptors, and molecular interactions.

NK cell receptors		Corresponding ligands	Function	Ref.
Activating receptors	CD16 (FcγRIIIa)	Fc segment of IgG	Mediates ADCC through Fc receptor engagement, significantly potentiating tumor cell elimination	[22]
	NKG2D (CD314)	MICA/B. ULBP1-6	Identifies stress-induced ligands on cancerous/infected cells, initiating cytotoxic responses and IFN-γ production	[25,26]
	NKp46 (CD335)	Viral hemagglutinin. Tumor-associated glycoproteins	Mediates targeted cytolysis of virus-infected cells while playing a pivotal role in antitumor immunosurveillance through direct cytotoxic activity	[27,28]
	NKp30 (CD337)	B7-H6. BAT3	Modulates dendritic cell maturation process while potentiating antitumor cytotoxic activity through enhanced immune recognition	[29]
	NKp44 (CD336)	Tumor-associated glycoproteins	Predominantly expressed in activated NK cells, this molecule significantly potentiates the specific cytolytic activity against solid malignancies	[30]
	DNAM-1 (CD226)	CD112 (PVR). CD155 (NECTIN-2)	Forms a corecognition complex with NKG2D to detect MHC-I-deficient cancer cells, initiating perforin-dependent cytolytic pathways for their selective elimination	[31,32]
Inhibitory receptors	KIR family	MHC-I molecules (HLA-A/B/C)	Recognize autologous MHC-I molecules to deliver inhibitory signals that maintain immune tolerance	[33,34]
	NKG2A (CD159a)	HLA-E (binds to MHC-I-derived peptide segments)	Suppresses NK cell activation and preserves self-tolerance through inhibitory receptor signaling	[33,35]
	TIGIT	CD155 (NECTIN-2)	Competes with DNAM-1 for shared ligand binding, thereby attenuating its tumor-suppressive signaling cascade	[31]

NK: Natural killer; IgG: Immunoglobulin G; ADCC: Antibody-dependent cellular cytotoxicity; MICA/B: Major histocompatibility complex class I chain-related proteins A and B; ULBP: UL16-binding protein; IFN-γ: Interferon-gamma; BAT3: Human leukocyte antigen-B-associated transcript 3; PVR: Poliovirus receptor; NKG2D: Natural killer group 2, member D; KIR: Killer cell immunoglobulin like receptor; MHC-I: Major histocompatibility complex class I; HLA: Human leukocyte antigen; TIGIT: Tyrosine-based inhibitory motif domain; DNAM-1: DNAX accessory molecule-1.

NK CELL DYSFUNCTION IN THE COLORECTAL TUMOR MICROENVIRONMENT

NK cells have emerged as critical mediators of antitumor immunity in CRC and have demonstrated a remarkable capacity for immune surveillance through their ability to detect and eliminate transformed cells[1,39]. These innate lymphocytes recognize tumor cells that exhibit altered major histocompatibility complex (MHC) class I expression, a phenomenon known as “missing-self” recognition, while they simultaneously respond to stress-induced ligands through activating receptors such as NKG2D and DNAM-1[16,17,40]. Clinical evidence increasingly supports the prognostic value of tumor-infiltrating NK cells, as their gene expression signatures reflect their activity (e.g., natural cytotoxicity-triggering receptor 3 and KLRC1), which is significantly correlated with improved patient outcomes; NK cells also serve as potential biomarkers for predicting the response to immunotherapy[1,41]. The CRC microenvironment, however, represents a formidable barrier to NK cell function, and is characterized by intricate cellular networks and immunosuppressive molecular cues[42-45]. The immunologically hostile niche contains not only malignant epithelial cells but also diverse immune populations, including tumor-associated macrophages and myeloid-derived suppressor cells (MDSCs), along with stromal components such as cancer-associated fibroblasts (CAFs) and extracellular matrix proteins[46,47]. Within this complex immunological ecosystem, NK cells encounter multifaceted functional suppression across multiple regulatory layers, as will be systematically discussed in subsequent sections (Figure 2)[45,48].

Figure 2 Suppressive mechanisms in the tumor microenvironment that impair natural killer cell function. The antitumor activity of natural killer (NK) cells is compromised in the colorectal cancer microenvironment through multiple inhibitory mechanisms[49-53,55,57-65,70,77-80]: (1) Immunosuppressive cytokines: Cytokines such as interleukin (IL)-8, IL-6, transforming growth factor-β, and IL-10, which are secreted by cancer-associated fibroblasts, M2 macrophages, and colorectal cancer cells, directly or indirectly suppress NK cell cytotoxicity and effector functions; (2) Immune checkpoint signaling: Inhibitory immune checkpoint pathways, including like tyrosine-based inhibitory motif domain, programmed cell death protein 1, NK group 2, member A, and signal-regulatory protein alpha, along with their respective ligands, transmit suppressive signals that dampen NK cell activation and promote immune evasion; and (3) Metabolic dysregulation: Mitochondrial dysfunction, oxidative stress and apoptosis in NK cells are exacerbated by mitochondrial impairment. Another mechanism is lactate-mediated suppression, where high lactate levels in the tumor microenvironment induce mitochondrial stress, which further inhibits NK cell activity. Finally, lipid accumulation due to aberrant lipid metabolism is indicated by elevated macrophage scavenger receptor 1, CD36, and CD68, which impairs NK cell function and reduces tumor-killing capacity. NK: Natural killer; IL: Interleukin; CAF: Cancer-associated fibroblast; CRC: Colorectal cancer; MSR1: Macrophage scavenger receptor 1; TGF: Transforming growth factor; Treg: Regulatory T cell; MDSC: Myeloid-derived suppressor cell; SIRPα: Signal-regulatory protein alpha; PD-L1: Programmed death ligand 1; NKG2A: Natural killer group 2, member A; PD-1: Programmed cell death protein 1; HLA: Human leukocyte antigen; TIGIT: Tyrosine-based inhibitory motif domain.

Inhibitory cytokines in the CRC microenvironment

The antitumor efficacy of NK cells in CRC is profoundly constrained by an array of inhibitory cytokines within the tumor microenvironment (TME). Transforming growth factor-beta (TGF-β), which is found at elevated levels on the peripheral blood mononuclear cells of CRC patients, exerts multifaceted immunosuppressive effects by both reducing IFN-γ production and downregulating the critical activating receptors NKG2D and DNAM-1 on NK cells[49]. This cytokine further establishes an immunosuppressive network through its synergistic action with IL-12, which upregulates NKG2A expression in CD8+ T cells[50]. Similarly, IL-10 demonstrates potent NK cell suppressive activity in the TME and impairs both cellular activation and cytotoxic potential[51], but also contributes to the increased NK cell-to-regulatory T-cell (Treg) ratio observed in the blood of CRC patients[52]. Moreover, the proinflammatory cytokine IL-6 has been shown to functionally hinder NK cell participation in the TME, thereby facilitating cancer progression and metastasis, as demonstrated in preclinical CRC models[53]. Intriguingly, IL-37, an anti-inflammatory cytokine that belongs to the IL-1 family, normally increases NK cell activity through IL-1 receptor 8 degradation; however, its significant downregulation in CRC tumor tissue is correlated with poorer patient outcomes and sustained NK cell dysfunction[54].

The immunosuppressive landscape is further complicated by tumor-associated macrophages and CAFs[55,56]. While STING-activated macrophages can transiently support NK cell function through IL-18/IL-1β-mediated 4-1BB/4-1BBL induction, their net effect remains predominantly suppressive[42]. CAFs orchestrate immune evasion through two mechanisms: Vascular cellular adhesion molecule-1-mediated facilitation of tumor cell-monocyte adhesion and IL-8-driven M2 macrophage polarization, which inhibits NK cell functions[55]. This collaborative suppression is compounded by the finding that CD38+ NK cell-derived factors can similarly promote M2 macrophage differentiation, establishing a vicious cycle of immune inhibition[52]. Additionally, MDSCs, which constitute one of the predominant immune cell populations in the CRC stroma, play a pivotal role in promoting tumor invasion and metastasis through their potent immunosuppressive activity[57]. Multiple findings of evidence demonstrate that MDSCs effectively suppress the proliferation and cytotoxic function of NK cells by the secretion of immunosuppressive mediators, including TGF-β and arginase-1[57,58]. In addition to their direct inhibition of NK cells, MDSCs establish a comprehensive immunosuppressive network by the disruption of antigen presentation and the facilitation of Treg expansion[58,59]. Within the CRC TME, Tregs have emerged as another critical negative regulator of NK cell function[58]. These cells employ both contact-dependent and contact-independent mechanisms to impair NK cell activity: (1) Through the production of inhibitory cytokines (IL-10 and TGF-β) that suppress the release of cytotoxic granules (perforin and granzyme B); and (2) Via competitive consumption of IL-2, thereby limiting the availability of this crucial cytokine for NK cell activation[58]. These findings collectively illustrate the complex cytokine-mediated mechanisms that subvert NK cell antitumor immunity in CRC.

Immune checkpoint regulation of NK cell function in CRC

The functional activity of NK cells in CRC is critically regulated by multiple immune checkpoint pathways that modulate their antitumor responses. These regulatory mechanisms involve both inhibitory and activating receptor-ligand interactions that fine-tune NK cell effector functions within the TME[43]. A key inhibitory axis is the PD-1/PD-L1 pathway, where tumor cell-expressed PD-L1 engages PD-1 on NK cells to suppress their cytotoxic potential, leading to weakened immune surveillance[60,61]. Clinical studies have demonstrated that PD-1+ NK cells exhibit impaired degranulation and cytokine production, which are correlated with reduced tumor infiltration and poorer patient outcomes[61]. Importantly, the percentage of PD-1+ NK cells increase with disease progression, suggesting that this pathway contributes to immune evasion in advanced CRC. The CD94/NKG2A-human leukocyte antigen (HLA)-E axis is another critical checkpoint in CRC[62]. NKG2A+ NK cells show significantly reduced effector function upon engagement with HLA-E expressed on tumor cells[63]. Notably, HLA-E expression is frequently upregulated in CRC samples, and its upregulation is associated with tumor cell resistance to NK-mediated killing[62,64]. This immunosuppressive interaction may explain the observed correlation between high NKG2A expression on tumor-infiltrating lymphocytes and diminished clinical response to immunotherapy.

Tyrosine-based inhibitory motif domain (TIGIT) is another immune checkpoint molecule widely expressed on NK cells, activated T cells, and Tregs[65]. Emerging evidence highlights the importance of the TIGIT-CD155 pathway in regulating NK cell activity in CRC[66]. TIGIT engagement by tumor-associated CD155 not only directly inhibits NK cell cytotoxicity but also competes with the activating receptor DNAM-1 for CD155 binding[67,68]. Preclinical models have demonstrated that TIGIT blockade promotes IFN-γ production by both NK and CD8+ T cells, revealing that this pathway may be a promising therapeutic target[65,69]. The CD47-signal-regulatory protein alpha (SIRPα) axis plays dual roles in CRC immunobiology. While primarily known for its ability to inhibit macrophage phagocytosis[70], CD47 expression on tumor cells also modulates NK cell function through interactions with SIRPα. Compared with healthy controls, postoperative CRC patients exhibit decreased NK cell numbers and IFN-γ levels, potentially reflecting CD47-mediated immune suppression[70]. However, the precise mechanisms underlying this regulation require further investigation in large patient cohorts.

These checkpoint interactions collectively create an immunosuppressive network that limits NK cell efficacy in CRC. Understanding these regulatory pathways provides critical insights for developing combination immunotherapies that can restore NK cell antitumor activity while overcoming microenvironmental suppression. Current therapeutic strategies targeting these checkpoints, including PD-1/PD-L1 blockade and NKG2A inhibition, show promise in preclinical CRC models and warrant further clinical evaluation[35,45,71,72].

Metabolic reprogramming and NK cell dysfunction in CRC

The antitumor efficacy of NK cells in CRC is profoundly influenced by metabolic alterations within the TME[73]. Although the infiltration of a large number of NK cells is correlated with favorable clinical outcomes[74], compared with their peripheral blood counterparts, tumor-associated NK cells in CRC patients exhibit significant functional impairments characterized by reduced cytokine production and diminished degranulation capacity[75]. CAFs play a pivotal role in creating this immunosuppressive metabolic niche. Through stromal remodeling and proangiogenic activity, CAFs establish a microenvironment that not only supports tumor progression but also actively suppresses NK cell function and reduces the number of NK cells, resulting in a decrease in the efficacy of immunotherapy[76]. This metabolic disruption manifests through multiple mechanisms, which are discussed below.

Lipid metabolic disorders: Postoperative NK cell dysfunction in CRC patients has been linked to abnormal lipid accumulation[77]. Elevated expression of lipid scavenger receptors (macrophage scavenger receptor 1), CD36 and CD68 in lipid-rich NK cells is correlated with impaired tumoricidal capacity[77]. This metabolic reprogramming may explain the observed clinical association between obesity, a state of chronic metabolic dysregulation, and both increased cancer risk and altered NK cell phenotypes with diminished cytotoxic effects on cancer cells[78].

Mitochondrial impairment: NK cell functionality is intimately tied to mitochondrial fitness. In primary gastrointestinal cancers, mitochondrial dysfunction in NK cells is associated with oxidative stress and increased apoptosis[79]. Moreover, lactate accumulation in the TME exacerbates NK cell mitochondrial dysfunction, leading to the induction of mitochondrial stress and compromising NK cell surveillance capabilities[80]. This metabolic byproduct of tumor glycolysis creates a feed-forward loop of immune suppression.

These metabolic barriers collectively contribute to the dysfunction of NK cells in CRC. The resulting immunosuppression presents a significant challenge to immunotherapy efficacy, as metabolically impaired NK cells demonstrate reduced migration, activation, and tumor-killing capacity[36,81,82]. Currently, metabolic interventions, including pharmacological modulation of lipid metabolism, lactate dehydrogenase inhibition to counteract acidosis, and mitochondrial-targeted therapies to restore NK cell bioenergetics are being explored[39,83]. Understanding these complex metabolic interactions provides critical insights for developing strategies to overcome NK cell dysfunction and improve immunotherapeutic outcomes in CRC. The metabolic vulnerabilities of tumor-associated NK cells represent promising therapeutic targets for reversing immune suppression and restoring antitumor immunity.

EMERGING FRONTIERS IN NK CELL-BASED IMMUNOTHERAPY FOR CRC

Recent advances in cancer immunotherapy have highlighted NK cells as promising therapeutic agents for CRC due to their innate capacity for tumor recognition and potent cytotoxic effector functions[84-86]. Unlike T-cell-based approaches, NK cells exert antitumor effects without prior antigen sensitization and exhibit a favorable safety profile, which minimizes the risk of cytokine release syndrome and graft-vs-host disease[16]. In recent years, the application prospects of NK cells in the treatment of CRC have received increased attention. Due to their strong immune surveillance ability and direct cytotoxic functions, these cells are important candidates for cancer immunotherapy. Figure 3 shows the general generation and mechanism of chimeric antigen receptor (CAR)-NK cells in CRC therapy. The following are the main NK cell-related treatment methods that have been developed thus far.

Figure 3 Generation and mechanism of chimeric antigen receptor-natural killer cells in colorectal cancer therapy[16,131-135,152,153,159,160]. Cell sources for chimeric antigen receptor (CAR)-natural killer (NK) generation: CAR-NK cells can be derived from multiple sources, including peripheral blood mononuclear cells from healthy donors, umbilical cord blood, hematopoietic progenitor cells differentiated from induced pluripotent stem cells, and the immortalized NK-92 cell line. Genetic modification methods: Two primary approaches are used for CAR delivery, including nonviral methods (electroporation for transient mRNA transfection), and viral methods (retroviral or lentiviral vectors for stable DNA integration). Cell processing and expansion: Following genetic modification, the cells undergo ex vivo expansion to achieve therapeutic quantities as well as radiation treatment (for certain cell types derived from NK-92) to ensure safety. Therapeutic mechanism in colorectal cancer: When infused into patients, CAR-NK cells recognize and bind to tumor-associated antigens on colorectal cancer cells, and initiate cytotoxic responses through the release of perforin and granules containing perforin and granzyme B and the secretion of pro-inflammatory cytokines, thereby effectively eliminating tumor cells while minimizing off-target effects. PBMC: Peripheral blood mononuclear cell; UCB: Umbilical cord blood; NK: Natural killer; CAR: Chimeric antigen receptor; iPSC: Induced pluripotent stem cell; HPC: Hematopoietic progenitor cell; GZMB: Granules containing perforin and granzyme B; CRC: Colorectal cancer; IFN-γ: Interferon-gamma; TNF-α: Tumor necrosis factor-α.

Cytokine-based therapeutic strategies to activate NK cells in CRC

Cytokines play pivotal roles in modulating NK cell function and have emerged as key therapeutic agents in CRC immunotherapy. Among these, IL-2 and IL-15 have demonstrated significant potential in promoting NK cell proliferation and cytotoxicity, although their clinical application presents distinct challenges[87-90].

IL-2: IL-2 was among the first cytokines used to stimulate NK cell activity and can expand NK cell populations and augment antitumor responses[91]. However, IL-2 therapy is limited by its systemic toxicity stemming from its broad immune activation ability, and this toxicity can result in cytokine release syndrome and vascular leak syndrome[92,93]. For example, the concomitant expansion of Tregs may promote immunosuppression in the TME[94,95]. To address this, combination strategies, such as pairing IL-2 with cetuximab [anti-epidermal growth factor receptor (EGFR)] to stimulate NK cell activity, thereby enhancing ADCC, have been explored[64]. Recent innovations, such as IL-2 mimetics (e.g., Neo-2/15) engineered to selectively engage IL-2 receptor-beta/gamma (IL-2Rβγ) while avoiding IL-2Rα and IL-15Rα binding, have shown improved efficacy and reduced toxicity in preclinical CRC models, offering a promising path forwards[96]. This strategy has significant potential for both signaling protein modulation and the development of high-precision therapeutic agents[96].

IL-15: In contrast to IL-2, IL-15 potently activates NK cells without stimulating Tregs[87,97], and thus it is an attractive candidate for CRC immunotherapy. Emerging evidence has demonstrated that IL-15 can effectively counteract the immunosuppressive effects of TME-derived soluble factors on NK cells. Through metabolic programming, IL-15 restores mitochondrial function and bioenergetic capacity in NK cells, thereby rescuing their cytotoxic activity against tumor targets[81]. To optimize its therapeutic potential, an engineered IL-15 variant, a fusion protein composed of an attenuated IL-15 mutant and an anti-PD-1 antibody (anti-PD1-IL15m), has been developed; this infusion protein enhances tumor-targeted delivery and reduces systemic immune activation[98]. Additionally, IL-15 has been employed to augment the cytotoxicity of CAR-NK cells to tumor cells and CAFs by targeting CD70[99]. Notably, efbalropendekin alfa (XmAb24306), an IL-15/IL15Rα-Fc fusion protein engineered with reduced CD122-binding affinity for IL-15, exhibits both prolonged therapeutic efficacy and reduced systemic toxicity[96]. This modified cytokine significantly enhances NK cell-mediated cytotoxicity to CRC cell lines and patient-derived three-dimensional tumor spheroids[100].

Emerging cytokine combinations and delivery strategies: In addition to IL-2 and IL-15, novel cytokine-based approaches are under investigation. For example, a decoy-resistant IL-18 mutant (DR18), produced by nonpathogenic Escherichia coli, has demonstrated potent antitumor effects in immunocompetent CRC models by activating both NK and CD8+ T cells[101]. Similarly, intratumoral administration of adenovirus encoding DR18 and IL-12 synergistically enhances NK and T-cell infiltration, and suppresses the growth of both local and distal tumors, which highlights the potential of localized cytokine delivery to minimize systemic toxicity while maximizing immune activation[102].

Despite these advances, cytokine therapy faces several hurdles, such as dosing optimization, TME immunosuppression and precision deliver. Ongoing clinical trials and preclinical studies continue to refine cytokine-based regimens, with the goal of integrating these agents into multimodal therapies for CRC. By addressing current limitations, next-generation cytokine therapies may unlock durable NK cell-mediated antitumor responses in CRC patients.

Immune checkpoint inhibition to augment NK cell-mediated antitumor immunity in CRC

The strategic blockade of immune checkpoint pathways has emerged as a promising approach to reinvigorate NK cell function in CRC. These inhibitory receptors, which normally maintain immune homeostasis, are frequently coopted by tumor cells to evade immune surveillance. Targeted inhibition of these pathways can effectively restore NK cell cytotoxicity and enhance antitumor responses through multiple mechanisms.

The PD-1/PD-L1 axis: The PD-1/PD-L1 pathway serves as a pivotal immune checkpoint in CRC, where PD-L1 expression on tumor cells directly inhibits NK cell activity via PD-1 engagement[103]. Disruption of this signaling axis restores NK cell effector functions, significantly enhancing tumor cell elimination efficiency[60,71,104,105]. Thus, PD-1/PD-L1 antibodies may reinvigorate the antitumor activity of NK cells by blocking immunosuppressive signals in the TME, demonstrating measurable efficacy in CRC[106]. Specifically, preclinical studies have demonstrated that PD-1 blockade in xenograft models induces substantial tumor regression, an effect completely abolished upon NK cell depletion, highlighting the critical role of NK cells in mediating this therapeutic response[107]. Clinically, the PD-1/PD-L1+ TME results in impaired degranulation and IFN-γ production, suggesting that checkpoint inhibition may reverse this functional exhaustion[108-110].

Novel checkpoint targets beyond PD-1: Emerging strategies focus on additional inhibitory receptors that regulate NK cell activity in CRC. For example, the inhibition of the NKG2A-HLA-E axis with the anti-NKG2A monoclonal antibody, monalizumab, can promote NK cell degranulation and tumor cell killing across multiple malignancies, including CRC, with particular promise when combined with existing PD-1/PD-L1 blockade strategies[62,63]. While primarily characterized in T cells, combinatorial approaches pairing cytotoxic-T-lymphocyte-associated antigen 4 blockade with IL-2 have shown synergistic effects in augmenting NK cell function, suggesting the potential for multimodal immunotherapy regimens[111]. Moreover, TIGIT, an immunoregulatory receptor expressed on activated NK cells, mediates functional suppression through engagement with its ligand CD155 on tumor cells[66]. Preclinical studies have demonstrated that TIGIT inhibition prevents NK cell dysfunction while promoting cytotoxic granule release and IFN-γ production, thereby restoring antitumor immunity[112,113]. Notably, TIGIT is frequently co-expressed with the complementary checkpoint poliovirus receptor-related immunoglobulin domain protein on tumor-infiltrating lymphocytes, creating a coordinated immunosuppressive network. This biological insight has led to the development of a novel bispecific antibody that targets both pathways, which, in early clinical trials, increased NK cell cytotoxicity by 1.8-fold across multiple cancer types, including CRC[114]. The therapeutic potential of checkpoint modulation is further highlighted by observations that conventional radiotherapy paradoxically upregulates TIGIT expression on NK cells while downregulating functional markers such as CD107a, granzyme B and IFN-γ, effects that can be effectively counteracted through combined TIGIT blockade[115]. Taken together, these findings underscore the importance of multiple checkpoint-targeting approaches to overcome NK cell exhaustion in the TME while providing a strong rationale for combination strategies that integrate checkpoint inhibition with conventional antitumor therapies. Current clinical efforts continue to refine these approaches, with a particular focus on the optimization of therapeutic sequencing and the identification of predictive biomarkers of treatment response.

ADCC in CRC immunotherapy

NK cells play a pivotal role in mediating ADCC, a critical mechanism by which monoclonal antibodies exert their antitumor effects. This process is initiated when the Fab domain of tumor-targeting antibodies binds to specific antigens on cancer cells, while their Fc domain engages CD16A (Fc region receptor IIIA) receptors on NK cells, triggering potent cytotoxic responses[116]. Harnessing the ADCC mechanism has demonstrated significant clinical success, particularly in hematological malignancies[117], and the efficacy of therapies that induce ADCC is currently being explored in solid tumors, including CRC, as summarized in Table 2[118-125].

Table 2 Antibody-dependent cellular cytotoxicity-inducing therapeutic antibodies in colorectal cancers.

Antibody	Mechanism	Ref.
GA201	Anti-EGFR mAb engineered to enhance ADCC, thereby potentiating NK cell-mediated cancer cell lysis	[118]
Cetuximab	EGFR-targeted therapeutic agent that mediates NK cell-dependent ADCC, currently approved as first-line treatment for mCRC	[119,120]
hPR1A3	Demonstrates a 10-fold enhancement in ADCC when evaluated using NK effector cells against CEA-expressing CRC cells	[121]
Trastuzumab	Potentiates NK cell-mediated ADCC in combination with lenalidomide, irrespective of KRAS mutation status or FcγRIIIa polymorphism	[122]
mAb CC4	Enhances NK cell cytotoxic activity against MHC-I-deficient CRC cells	[123]
CO17-1A	NK cell stimulatory factor significantly enhances the cytolytic activity of NK cells against CRC cells	[124]
AFM24	Demonstrates efficacy against EGFR-expressing tumors independent of EGFR expression levels or KRAS/BRAF mutation status	[125]

EGFR: Epidermal growth factor receptor; ADCC: Antibody-dependent cellular cytotoxicity; NK: Natural killer; mCRC: Metastatic colorectal cancer; CEA: Carcinoembryonic antigen; CRC: Colorectal cancer; FcγRIIIa: Fc region receptor IIIa; MHC-I: Major histocompatibility complex class 1.

In CRC, EGFR/human EGFR 1 (HER1) represents a particularly promising target for ADCC-based therapies. EGFR is overexpressed in approximately 75% of CRC cases and drives tumor progression by enhancing proliferation, survival, and metastatic potential[126]. Cetuximab, an anti-EGFR monoclonal antibody, has shown the capacity to harness NK cell-mediated ADCC against EGFR-expressing CRC cells[125,127,128]. Recent studies have further demonstrated that the combination of cetuximab and reovirus-activated NK cells results in the synergistic enhancement of antitumor cytotoxicity, and suggest novel combinatorial approaches for CRC clinical translation[129]. This strategy capitalizes on reovirus-induced NK cell activation and simultaneously engages ADCC through EGFR targeting, potentially overcoming the limitations of single-agent therapies.

These developments highlight the increasing recognition of ADCC as a crucial component of anti-CRC immunity and underscore the importance of optimizing both antibody design and NK cell function to maximize therapeutic efficacy. Current challenges include overcoming tumor heterogeneity in target antigen expression and addressing immunosuppressive mechanisms that may limit NK cell activity in the TME. Future directions may involve the development of bispecific antibodies that simultaneously target tumor antigens and NK cell activation receptors, as well as strategies to increase NK cell persistence and infiltration into CRC lesions[116,130].

Ex vivo expansion and adoptive transfer of engineered NK cells for CRC therapy

The adoptive transfer of NK cells, that have been expanded ex vivo and genetically modified, has emerged as a promising immunotherapeutic approach for CRC, as this modality offers several distinct advantages over conventional T-cell-based therapies[16,131]. Unlike antigen-specific T cells, NK cells mediate rapid antitumor responses through innate recognition mechanisms independent of MHC presentation; this enables them to effectively target tumor cells with downregulated MHC class I molecules, which is a common immune evasion strategy in CRC. This intrinsic capability, combined with their reduced risk of graft-vs-host disease and cytokine release syndrome, makes allogeneic NK cell products particularly attractive for off-the-shelf immunotherapy applications[16].

Recent advances in cell engineering have significantly enhanced the therapeutic potential of adoptive NK cell transfer[132]. For example, CARs are synthetic fusion proteins that contain three fundamental domains: An extracellular antigen-binding domain for target recognition, a transmembrane domain for structural stability, and an intracellular signaling domain for immune cell activation[16,133]. The continuous refinement of the CAR architecture has led to the development of four distinct generations, each representing progressive optimization through strategic modifications to the intracellular signaling domains for enhanced cell activation and prolonged persistence[16,133]. These engineered CAR-NK cells demonstrate potent activity against CRC in preclinical models and maintain favorable safety profiles[85,134,135]. Table 3 comprehensively summarizes recent preclinical progress in CAR-NK cell therapy development for CRC[99,136-143]. The data highlight multiple promising target antigens, including epithelial cell adhesion molecule (EpCAM), carcinoembryonic antigen (CEA), and HER2, with detailed mechanistic insights into their respective roles in promoting antitumor activity. To date, the clinical translation of CAR-NK-cell therapy for CRC remains limited, as few registered clinical trials are currently active. A phase I study (NCT05213195) investigating NKG2D-targeted CAR-NK cells in patients with treatment-refractory metastatic CRC is underway. This early-in-human trial aims to evaluate the safety and preliminary efficacy of this innovative immunotherapy approach. These collective findings strongly support the therapeutic potential and clinical translatability of CAR-NK cell technology for CRC treatment.

Table 3 Target antigens for chimeric antigen receptor-natural killer cell therapy in colorectal cancer preclinical studies.

Targets	Mechanism	Ref.
EpCAM	CAR-engineered NK-92 cells demonstrate specific recognition and potent cytotoxicity against EpCAM-expressing CRC cells, mediated through targeted release of effector molecules including IFN-γ, perforin and granzyme B	[136]
NKG2D ligands	The CAR construct incorporating NKG2D’s extracellular domain and DAP12 signaling module significantly enhances NK cell-mediated tumoricidal activity	[137]
CD70	CD70 represents an ideal therapeutic target in CRC due to its tumor-restricted expression profile. IL-15-aremed CAR-NK cells demonstrate potent elimination of CD70+ cancerous cells	[99]
CEA	Retrovirally transduced CEA-specific CAR-NK-92MI cells demonstrate significantly enhanced cytotoxic activity (2-3 fold increase) against CEA-expressing tumor models	[138,139]
MSLN	MSLN-directed CAR immunotherapy demonstrates potent efficacy against mesothelin-high CRC models, achieving > 80% tumor volume reduction	[140]
HER2	The HER2-targeted CAR-NK platform represents a novel therapeutic strategy for HER2-amplified CRC, a molecular subset occurring in approximately 5% of cases	[139]
EGFRvIII	EGFRvIII-specific CAR-NK-92 cells demonstrate potent cytotoxic activity against tumor organoids expressing this pan-cancer neoantigen, showing > 90% target cell elimination in preclinical evaluation	[141]
Frizzled receptors	Frizzled receptor-targeted CAR therapies demonstrate specific pro-apoptotic activity against the 15% of CRCs exhibiting Frizzled overexpression	[141]
CD133	CAR133-NK92 cells demonstrate specific cytotoxicity against CD133+ tumor cells through antigen-dependent recognition, while simultaneously synergizing with TLR5 agonist to activate host immune responses against antigen-negative (CD133-) tumor populations	[142]
CDH17	CDH17-CAR-NK cells exhibit potent and selective cytotoxicity against CDH17-high CRC cells. When combined with CD47 blockade, this approach demonstrates synergistic antitumor effects, resulting in enhanced tumor clearance	[143]

EpCAM: Epithelial cell adhesion molecule; CAR: Chimeric antigen receptor; NK: Natural killer; CRC: Colorectal cancer; IFN-γ: Interferon-gamma; NKG2D: Natural killer group 2, member D; DAP12: DNAX-activating protein of 12 kDa; IL: Interleukin; CEA: Carcinoembryonic antigen; MSLN: Mesothelin; HER2: Human epidermal growth factor receptor 2; EGFRvIII: Epidermal growth factor receptor variant III; TLR5: Toll-like receptor 5.

Innovative combinatorial approaches are further expanding the capabilities of adoptive NK cell therapy. A notable example is the dual-targeting strategy employing HER2-synNotch and CEA-CAR systems, which has shown remarkable accuracy and efficacy in HER2-amplified CRC by enabling logical gating of CAR-NK cell activation on the basis of tumor antigen combinations[139]. This approach significantly reduces off-target effects and improves the breadth of tumor recognition[139]. Additionally, novel activation methods using R848 (a toll-like receptor 7/8 agonist) have been shown to potently increase the NK cell-mediated lysis of CRC cells, which suggests the potential for ex vivo preconditioning protocols[144].

The field is also exploring innovative cytokine-support strategies to improve NK cell persistence and function in vivo. A breakthrough approach utilizes nonpathogenic Escherichia coli to deliver an engineered DR18, which has demonstrated an enhanced capacity to increase CAR-NK cell antitumor responses while minimizing systemic toxicity[101]. This localized cytokine delivery method addresses a key challenge in NK cell therapy by providing sustained activation signals within the TME without inducing harmful systemic inflammation[101].

Current clinical efforts are focused on optimizing: (1) Ex vivo expansion protocols to generate clinically relevant NK cell doses; (2) Genetic engineering strategies to increase tumor targeting and overcome immunosuppression; (3) Combination approaches consisting of cytokines, immune checkpoint inhibitors, or conventional therapies; and (4) Novel delivery systems for localized NK cell activation and persistence. These developments position NK-cell adoptive transfer as a versatile platform for CRC immunotherapy, and ongoing clinical trials are evaluating both allogeneic and engineered NK cell products. As the field progresses, key challenges include improvements in homing to solid tumors, validation of long-term persistence, and the development of reliable biomarkers to predict treatment response. The integration of advanced gene editing, synthetic biology approaches, and biomaterial-based delivery systems holds promise for overcoming these limitations and realizing the full potential of NK cell therapy for CRC patients.

OVERCOMING CHALLENGES AND ADVANCING FRONTIERS: NK CELL-BASED IMMUNOTHERAPY FOR CRC

Technical and manufacturing challenges in NK cell-based immunotherapy for CRC

The clinical translation of NK cell-based therapies for CRC faces several critical technical and manufacturing hurdles that impact both therapeutic efficacy and scalability. These challenges span from initial cell expansion to genetic modification and final product formulation, each presenting unique obstacles that must be addressed to realize the full potential of this treatment modality.

Ex vivo cell proliferation and functional preservation: A primary challenge lies in achieving robust ex vivo expansion of NK cells while preserving their functional integrity. In current employed protocols, sufficient cell number must be achieved while maintaining cytotoxic potency and persistence in vivo[145]. Recent insights into NK cell biology have identified distinct subsets with increased therapeutic potential, particularly FcRγ-deficient memory-like NK cells. Compared with conventional NK cells, these cells exhibit superior proliferative capacity, cytokine production, and resistance to apoptosis, making them attractive candidates for therapy[146-148]. Memory-like NK cell subsets are characterized by distinct transcriptional and functional profiles, notably exhibiting deficiencies in multiple critical transcription factors and signaling molecules, including the essential tyrosine kinase (spleen tyrosine kinase)[149]. These specialized NK cell populations demonstrate unique antiviral capabilities, particularly in the context of robust ADCC against virally infected targets such as human cytomegalovirus-infected cells and influenza virus-infected cells[149]. Notably, these subsets display a marked proliferative advantage in response to viral challenge, suggesting an adaptive-like expansion mechanism. Moreover, these memory-like NK cells can be generated through cytokine priming (IL-12/IL-15/IL-18) or coculture with engineered feeder cells (K562, PLH, and 221.AEH), demonstrating their enhanced and sustained responses against tumor targets[150,151]. For example, although metastatic CRC is associated with poor clinical outcomes, memory-like NK cells, which primarily function through the NKG2D receptor, exert significantly stronger cytotoxic effects on CRC cells than conventional NK cells do. Furthermore, these cells demonstrate enhanced therapeutic efficacy when used in combination with cetuximab[43]. However, standardized protocols for reliably producing these subsets at the clinical scale remain elusive.

Genetic engineering and manufacturing challenges in CAR-NK cell development: The genetic modification of NK cells to express CARs represents a transformative approach in cancer immunotherapy. However, there are still significant technical hurdles that must be overcome to achieve clinical success with this therapeutic modality. Central to this challenge is the precise integration of CAR constructs into NK cell genomes to achieve stable, controlled expression while maintaining cellular fitness and function[152]. Conventional methods relying on random viral integration often result in inconsistent CAR expression levels and potential insertional mutagenesis, compromising both therapeutic efficacy and patient safety[153]. In response, innovative genome editing strategies have emerged, with CRISPR-Cas9-mediated site-specific integration into safe genomic harbor loci such as the GAPDH 3’ untranslated region demonstrating particular promise[153]. This targeted approach not only ensures more predictable transgene expression but also minimizes the risks of disrupting essential cellular genes, as evidenced in engineered NK-92MI cell lines showing consistent CAR expression and preserved cytotoxic activity[153].

The selection of optimal gene delivery systems is another critical consideration in CAR-NK cell manufacturing. While lentiviral vectors remain widely used for their ability to mediate long-term transgene expression, their application in NK cells faces inherent limitations, including typically low transduction efficiencies and complex production requirements[154]. Recent advances in viral vector engineering have yielded improved tools, such as baboon envelope pseudotyped lentiviral vectors, which exploit natural NK cell tropism to achieve significantly higher transduction rates than conventional vesicular stomatitis virus-G systems[155]. Parallel work has elucidated the importance of the NK cell activation state during transduction, and cytokine prestimulation markedly increases vector uptake and transgene expression[156,157]. However, persistent concerns regarding viral vector safety profiles, manufacturing costs, and scalability have spurred the development of nonviral alternatives[158]. For example, electroporation-based mRNA delivery offers rapid, high-efficiency transient expression suitable for certain applications, whereas emerging technologies such as lipid nanoparticles and transposon systems provide new avenues for stable genetic modification without viral components[159,160].

Future perspectives in NK cell-based immunotherapy for CRC

The evolving landscape of NK cell therapy presents transformative opportunities for CRC treatment, driven by innovative combinatorial approaches and precision engineering strategies. Building upon the inherent tumor recognition capabilities of NK cells, current research focuses on overcoming microenvironmental suppression while enhancing targeted cytotoxicity through multifaceted interventions.

A promising direction involves the rational combination of CAR-NK cells with conventional therapies. Clinical evidence supports this approach, as shown in a prospective randomized controlled trial in which IL-2-activated NK cells plus XELOX chemotherapy (capecitabine plus oxaliplatin) significantly improved progression-free and overall survival in postoperative CRC patients[161]. Preclinical studies further revealed that CAR-NK cells synergize with chemotherapeutic agents (e.g., capecitabine), where chemotherapy-induced immunogenic cell death enhances NK cell activation through upregulated stress ligands, which increases immune cell accuracy and intensity[161,162]. In addition to their cytotoxic effects, targeted agents also exert complementary effects. For example, the toll-like receptor 5 agonist CBLB502 empowers CAR133-NK92 cells through direct CAR-dependent cytotoxicity against CD133+ tumor populations and indirect immune-mediated clearance of CD133- cells via endogenous immune activation[142]. Similarly, cadherin-17-targeted CAR-NK cells exhibit enhanced therapeutic efficacy when combined with CV1 (a CD47-SIRPα axis inhibitor) in gastrointestinal tumor treatment by increasing M1 macrophage numbers and enhancing their overall activation[143]. In contrast, EpCAM-specific CAR-NK-92 cells synergize with regorafenib to suppress EpCAM+ CRC growth through the coordinated targeting of tumor signaling pathways and surface antigens[136]. Angiogenesis inhibitors (e.g., bevacizumab) further augment CAR-NK cell infiltration into tumor tissues by normalizing the tumor vasculature[163]. The integration of immune checkpoint blockade further increases these effects, as NKG2A and TIGIT inhibitors show particular promise in reversing NK cell exhaustion in CRC models[62,63,112]. Additionally, the combination of low-dose chemotherapy, radiotherapy and NK cell therapy for CRC treatment may induce synergistic cytotoxicity mediated through the NK cell receptors NKp30 and NKG2D, while pharmacological blockade of these receptors significantly attenuates the cellular degranulation capacity[164]. This combinatorial approach results in a marked reduction in tumor-related indicators, suppression of metastatic progression, and enhanced antitumor efficacy against primary lesions[164].

The frontier of NK cell therapy also involves the exploration of novel activation paradigms. A novel combinatorial immunotherapy platform that integrates astragaloside III, an innate immune activator, with chlorin e6-mediated photodynamic therapy has been developed. This dual-component system exerts multimodal antitumor effects, including in vitro activation of NK cell effector functions, significant suppression of tumor cell proliferation, and enhanced immune cell infiltration in murine tumor models[165]. Mechanistically, the formulation potentiates NK cell-mediated cytotoxicity by inducing their infiltration into tumors, which indicates promising therapeutic potential for solid tumors[165].

Technological innovations enable unprecedented precision in NK cell engineering. Advances in single-cell analytics may facilitate patient-specific NK cell product development, where receptor profiles are tailored to individual tumor antigen landscapes[166,167]. Nanotechnology platforms are leveraged to improve cell aggregation efficiency and to enhance tumor targeting[168], which is exemplified by hydrogel-based delivery systems that locally release methyltransferase-like 3 inhibitors such as STM2457 to modulate chemokine networks and promote CAR-NK cell recruitment[169]. Concurrently, biomaterial scaffolds that carry oncolytic viruses such as decorin-expressing adenoviruses (rAd.DCN) establish immunogenic niches that promote NK cell proliferation and increase their killing activity while directly lysing tumor cells, thereby inhibiting tumor growth[170].

Notably, three-dimensional CRC organoid models have emerged as powerful tools for studying tumor-immune interactions; however, their application in NK cell research remains relatively underexplored compared with that of other immune cell types. Nevertheless, these systems exhibit distinct advantages over traditional two-dimensional cultures as they faithfully recapitulate tissue architecture and pathophysiological features, thereby enabling a more accurate investigation of NK cell-mediated antitumor mechanisms[171]. Building on this platform, recent work has validated the predictive value of patient-derived organoids in the assessment of treatment response, particularly for stage IV CRC patients undergoing chemotherapy[172,173]. More importantly, advanced co-culture systems that integrate CRC organoids with stromal components such as CAFs have elucidated the microenvironmental regulation of NK cell function. For example, studies have demonstrated that CAFs within these reconstructed TME models secrete immunomodulatory factors such as Dickkopf 1, which dually regulate organoid proliferation and suppress NK cell activity; this finding highlights the utility of these models for dissecting complex cell-cell interactions[174,175]. Notably, the tumor-selective cytotoxicity of EpCAM-targeted CAR-NK cells was rigorously validated using patient-derived CRC organoids, which better recapitulate tumor heterogeneity and microenvironmental interactions than do conventional cell lines[141]. Although the current literature on organoid-NK cell co-cultures remains limited, these pioneering studies collectively underscore the potential of organoid platforms to serve as predictive tools for immunotherapy optimization, and thus they warrant further investigation in the context of NK cell-based therapies. Future NK cell immunotherapy for CRC will focus on the optimization of combinatorial strategies with conventional therapies and leveraging cutting-edge engineering approaches to enhance tumor targeting, overcome immunosuppression, and maximize therapeutic efficacy, ultimately paving the way for more effective and personalized treatment paradigms.

CONCLUSION

CRC remains a formidable global health challenge, and innovative therapeutic approaches are needed. NK cell-based immunotherapy has emerged as a promising strategy that exerts multifaceted antitumor effects through cytokine regulation, immune checkpoint modulation, ADCC induction, and adoptive cell transfer. While these modalities show considerable clinical potential, current limitations including treatment-related toxicity, technical constraints in cell manufacturing, and variable therapeutic efficacy, underscore the need for continued optimization. The field must address critical challenges in NK cell expansion protocols, functional persistence, and precision engineering of CAR-NK products. Emerging solutions that incorporate advanced gene-editing tools, biomaterial-enhanced delivery systems, and artificial intelligence-driven manufacturing processes are paving the way for more reliable and potent NK cell therapy. Future progress will likely stem from rationally designed combination regimens that synergize NK cell activation with microenvironment modulation, coupled with patient-specific approaches informed by multiomics profiling.

As these scientific and technological advances mature, NK cell-based immunotherapy is poised to transition from experimental treatment to clinical reality. Successful translation will require concerted efforts to standardize production, validate biomarkers, and optimize delivery strategies. With continued innovation, NK cell therapies may fundamentally transform the therapeutic landscape in CRC, particularly for patients with microsatellite-stable tumors that are resistant to conventional immunotherapies[176]. The coming decade promises to be transformative, as we overcome existing barriers and fully harness the therapeutic potential of these innate immune effectors.