Adoptive cell therapy in colorectal cancer: Advances in chimeric antigen receptor T cells

Abstract

Colorectal cancer (CRC) is the third most common cancer worldwide and remains a major treatment challenge, particularly in advanced and metastatic stages. Current standard treatments have limited efficacy, underscoring the urgent need for innovative strategies. Adoptive cell therapy (ACT), which involves in vitro expansion or genetic engineering of immune cells, is a promising approach to bolster anti-tumor immune responses. Key ACT modalities include chimeric antigen receptor (CAR) T cells, tumor-infiltrating lymphocytes (TILs), and T cell receptor (TCR)-engineered T cells. CAR-T cell therapy has shown success in hematological malignancies but faces significant challenges in solid tumors like CRC. These challenges include antigen heterogeneity, an immunosuppressive tumor microenvironment, on-target off-tumor toxicity, among other factors. To address these limitations, combinatorial approaches, such as immune checkpoint inhibitors, cytokines, and advanced gene-editing tools like CRISPR/Cas9, are being actively explored. These strategies aim to enhance CAR-T cell specificity, improve resistance to immunosuppressive signals, and optimize in vivo functionality. This review summarizes ACT approaches for CRC, with a focus on CAR-T therapy. It briefly introduces TILs and TCR-T cells, while emphasizing the major challenges faced by CAR-T therapy in solid tumors and discusses potential strategies to improve therapeutic outcomes.

Key Words: Colorectal cancer; Adoptive cell therapy; Immunotherapy; Chimeric antigen receptor T cells; Tumor-infiltrating lymphocytes; T-cell receptor-engineered T cells

Core Tip: This review discusses adoptive cell therapy approaches for colorectal cancer (CRC), emphasizing chimeric antigen receptor (CAR) T cell therapy. Despite its success in hematological malignancies, CAR-T therapy faces challenges in solid tumors like CRC, including antigen heterogeneity, tumor microenvironment immunosuppression, and on-target off-tumor toxicity. In this review, we explore combinatorial strategies, such as immune checkpoint inhibitors and CRISPR/Cas9 gene editing, to overcome these challenges and enhance CAR-T cell specificity, resistance to immunosuppressive signals, and in vivo functionality.

INTRODUCTION

Colorectal cancer (CRC) is one of the most common malignant tumors worldwide, ranking third in cancer incidence and second in cancer-related mortality[1]. According to GLOBOCAN Estimates in 2020, approximately 1.93 million new CRC cases and 935000 deaths were reported globally. By 2040, CRC incidence and mortality are expected to rise to 3.2 million and 1.6 million cases, respectively, although these may be further exacerbated by increasing risk factors associated with aging and environmental exposure[2]. Notably, patients with advanced and metastatic CRC, who account for most CRC-related deaths, have a dismal 5-year survival rate of less than 15%[3]. These findings underscore the urgent need to optimize existing diagnostic and therapeutic strategies for CRC.

Current standard treatments for CRC include surgical resection, chemotherapy (e.g., FOLFOX and FOLFIRI regimens), radiotherapy, and targeted therapies [e.g., anti-epidermal growth factor receptor (EGFR) and anti-vascular endothelial growth factor monoclonal antibodies] (Table 1)[4]. While these approaches show efficacy in localized CRC, their impact is significantly reduced in advanced and metastatic CRC[5]. For example, chemotherapy resistance is commonly linked to significant toxic side effects and the development of drug resistance. Moreover, EGFR inhibition is ineffective against colorectal tumors with over-activated KRAS signaling. Consequently, targeted therapy is less effective in CRC patients with KRAS mutations, which are present in around 40% of cases[6]. Furthermore, the CRC tumor microenvironment is complex and displays strong immunosuppressive features, including immune checkpoint molecule overexpression and inhibitory cytokine secretion, which further impair anti-tumor immunity[7]. Therefore, overcoming the limitations of conventional therapies and developing innovative treatment strategies have become critical areas of focus in current CRC research.

Table 1 Standard treatments for colorectal cancer.

Treatment method	Indication	Description
Surgical treatment	Early-stage colorectal cancer	Resection of the primary tumor along with regional lymph nodes, typically for localized tumors without distant metastasis
Chemotherapy	Early-stage and advanced colorectal cancer	Systemic treatment using cytotoxic agents to eradicate cancer cells. Chemotherapy is commonly employed adjuvantly after surgical resection to reduce recurrence risk or as a primary treatment in metastatic cases. Common agents include 5-FU, oxaliplatin, and irinotecan
Radiation therapy	Rectal cancer	Primarily utilized in rectal cancer for preoperative tumor downstaging or as postoperative adjuvant therapy to reduce the risk of local recurrence
Targeted therapy	Advanced colorectal cancer	Involves agents that specifically target cancer cell molecular markers, such as anti-angiogenic therapies or monoclonal antibodies. Bevacizumab (Avastin) and cetuximab (Erbitux) are among the most commonly used agents
Immunotherapy	Advanced colorectal cancer, particularly MSI-H tumors	Employs immune checkpoint inhibitors to enhance the body’s immune response against cancer. Pembrolizumab (Keytruda) and nivolumab (Opdivo) are commonly used in microsatellite instability-high tumors
Interventional therapy	Locally recurrent or unresectable advanced CRC	Includes localized procedures such as radiofrequency ablation and transarterial chemoembolization, aimed at controlling tumor progression in patients with advanced or inoperable disease
Chemotherapy combination	Advanced colorectal cancer, particularly metastatic CRC	A multimodal approach combining chemotherapy with targeted therapy or immunotherapy to improve therapeutic efficacy, commonly using regimens like FOLFOX or FOLFIRI

CRC: Colorectal cancer; MSI-H: Microsatellite instability-high; 5-FU: 5-fluorouracil.

Adoptive cell therapy (ACT) is an immunotherapeutic strategy that involves the ex vivo enhancement of antitumor functions in patient-derived or donor T cells, followed by reinfusion into the patient. ACT has offered additional treatment options for various solid tumors, including CRC. The primary forms of ACT include chimeric antigen receptor T-cell (CAR-T) cells, tumor-infiltrating lymphocytes (TILs), and T cell receptor (TCR)-engineered T cells.

KEY ACT MODALITIES

CAR-T cells

CAR-T cell therapy starts by collecting T cells from the patient. These cells are then genetically engineered to express a receptor that targets specific tumor antigens. First, the modified CAR-T cells are expanded ex vivo. After lymphodepleting chemotherapy, the CAR-T cells are infused back into the patient. Once infused, the CAR-T cells recognize and eliminate tumor cells that express the target antigen (Figure 1). A key advantage of CAR-T therapy is its independence from major histocompatibility complex (MHC) molecules, effectively overcoming T cell dysfunction caused by tumor-induced MHC downregulation[8,9]. Despite achieving groundbreaking success in hematologic malignancies[10], the efficacy of CAR-T cells in CRC has been significantly limited by the immunosuppressive tumor microenvironment and antigen heterogeneity (Table 2). To address these challenges, researchers have been developing CAR-T cells targeting CRC-associated antigens such as epithelial cell adhesion molecule (EpCAM), carcinoembryonic antigen (CEA), cellular-mesenchymal epithelial transition (c-Met), and tumor-associated glycoprotein 72[11]. In addition, other studies focus on enhancing tumor infiltration, such as combining CAR-T therapy with vascular disruptive agents like combretastatin A-4 phosphate (CA4P)[12,13]. Moreover, the combination of immune checkpoint inhibitors and other tumor microenvironment-modulating strategies is considered a promising strategy to augment CAR-T cell efficacy[14]. While initial clinical trials have shown that CAR-T therapy is safe and has early signs of efficacy in CRC, its widespread application remains hindered by critical challenges, including off-target effects, tumor microenvironment-mediated suppression, and antigen heterogeneity.

Figure 1 Schematic diagram of chimeric antigen receptor-T cell construction and preparation. Chimeric antigen receptor (CAR)-T therapy involves isolating T cells from the patient’s blood, genetically engineering them to express CARs, and expanding the modified cells in vitro. After lymphodepleting therapy, the engineered CAR-T cells are infused back into the patient, where they recognize and eliminate tumor cells. PBMC: Peripheral blood mononuclear cell; CAR: Chimeric antigen receptor; scFv: Single-chain variable fragments; VL: Variable region of light chain; VH: Variable region of heavy chain. (Created with BioRender.com).

Table 2 Chimeric antigen receptor-T drugs approved by the Food and Drug Administration and China.

Drug name	Indication	Approval date	Developer	Approved countries	Target
Kymriah (tisagenlecleucel)	B-cell acute lymphoblastic leukemia in children and young adults	August 30, 2017 (FDA)	Novartis	United States, European, Canada, Australia, China	CD19
		December 2018 (China)
Yescarta (axicabtagene ciloleucel)	Large B-cell lymphoma in adults	October 18, 2017 (FDA)	Kite Pharma (Gilead)	United States, European, Canada, China	CD19
		February 2020 (China)
Tecartus (brexucabtagene autoleucel)	Relapsed or refractory mantle cell lymphoma	July 24, 2020 (FDA)	Kite Pharma (Gilead)	United States, European, etc.	CD19
Breyanzi (lisocabtagene maraleucel)	Large B-cell lymphoma in adults	February 5, 2021 (FDA)	Bristol Myers Squibb	United States, European, Canada, etc.	CD19
Abecma (idecabtagene vicleucel)	Multiple myeloma	March 26, 2021 (FDA)	Bristol Myers Squibb/Bluebird Bio	United States, European, China	BCMA
		February 2022 (China)
Carvykti (ciltacabtagene autoleucel)	Multiple myeloma	February 28, 2022 (FDA)	Johnson and Johnson/Legend Biotech	United States, European, China	BCMA
		November 2022 (China)
Kymriah (tisagenlecleucel)	Large B-cell lymphoma in adults	May 2021 (China)	Novartis	China	CD19
Blinatumomab (blincyto)	B-cell acute lymphoblastic leukemia	October 2019 (China)	Amgen	China	CD19

CD: Cluster of differentiation; BCMA: B-cell maturation antigen; FDA: Food and Drug Administration.

TIL therapy

TIL therapy adopts lymphocytes from the patient’s tumor tissue, expands tumor-reactive TILs ex vivo, and reinfuses them into the patient to enhance antitumor immune responses. The process commences with the division of the tumor tissue into small pieces, followed by digestion. TILs are subsequently cultured with interleukin (IL)-2, then subsequently expanded for 3-5 weeks. The enzyme linked immune spot assay is used to assess the ability of TILs to recognize tumor antigens and release cytokines like interferon (IFN)-γ[15]. After initial amplification and selection of tumor-specific TILs (preREP), the rapid expansion phase begins. TILs can be further expanded by using activating antibodies [e.g., OKT3, anti-cluster of differentiation (CD) 28 antibodies] or binding to magnetic beads. After expansion, TILs are cryopreserved and transferred to designated treatment centers. Before infusion, patients undergo non-myeloablative chemotherapy to deplete suppressive immune cells. In addition, IL-2 is administered post-infusion to stimulate TIL proliferation and activation[16] (Figure 2). While TIL therapy has shown potential in CRC patients with high microsatellite instability (MSI-H)[17], its efficacy is constrained by tumor antigen heterogeneity and the immunosuppressive tumor microenvironment[18]. However, recent studies suggest that optimization of TIL isolation and expansion techniques, combined with anti-programmed cell death protein 1 (PD-1) monoclonal antibodies, can significantly enhance therapeutic outcomes[19].

Figure 2 Schematic diagram of tumor-infiltrating lymphocytes construction and preparation. The process includes tumor digestion and tumor-infiltrating lymphocyte (TIL) isolation, interleukin (IL)-2-mediated culture and expansion, tumor antigen recognition assessment via enzyme linked immune spot assay, and rapid expansion using activating antibodies or magnetic beads. Expanded TILs are cryopreserved and transferred to treatment centers. Before infusion, patients receive lymphodepleting chemotherapy, followed by IL-2 administration to stimulate TIL proliferation and activation. TIL: Tumor-infiltrating lymphocyte; IL: Interleukin; ELISPOT: Enzyme-linked immune spot assay; CD: Cluster of differentiation. (Created with BioRender.com).

TCR-engineered T cells

TCR therapy involves genetically engineering T cells to express TCRs that specifically recognize tumor antigens presented by human leukocyte antigen (HLA) molecules, thereby enhancing tumor-specific immune responses[20]. First, tumor samples and peripheral blood are collected from the patient. Tumor-specific antigens (TSA) are then identified through proteomics (mass spectrometry) and genomic sequencing (whole exome and RNA sequencing). The selected tumor antigen peptides are subsequently used to stimulate and expand patient-derived T cells. HLA-I tetramer staining is employed to select T cells that specifically recognize the antigen. TCR gene sequencing is performed to obtain the TCR-α and TCR-β chain sequences. These TCRs are then cloned and tested for functionality and safety[21]. Finally, the specific TCR genes are introduced into the patient’s peripheral blood T cells using lentiviral transfection or other genetic engineering methods. The engineered TCR-T cells are generated, expanded, and ultimately infused back into the patient to specifically recognize and attack tumor cells through the immune system[22] (Figure 3).

Figure 3 Schematic diagram of T-cell receptor-T cell construction and preparation. T-cell receptor (TCR)-T cell therapy involves genetically modifying T cells to express tumor-specific TCRs, enhancing immune recognition of tumor antigens presented by human leukocyte antigen (HLA) molecules. The process includes tumor and blood sample collection, antigen identification via proteomics and sequencing, T cell stimulation and selection using HLA-I tetramer staining, TCR gene sequencing and cloning, and lentiviral transfection to generate engineered TCR-T cells. These cells are expanded and reinfused into the patient to specifically target and eliminate tumor cells. TCR: T-cell receptor; IFN: Interferon; TNF: Tumor necrosis factor; APC: Antigen-presenting cell; PBMC: Peripheral blood mononuclear cell. (Created with BioRender.com).

Studies targeting KRAS mutations and MSI-H CRC have demonstrated promising efficacy of TCR therapy in these subtypes[5]. Furthermore, a recent phase 2 clinical trial evaluated the adoptive transfer ACT of TCR-transduced T cells targeting personalized neoantigens in metastatic non-MSI-H CRC. Results showed durable regression of metastatic tumors lasting 4 to 7 months in some patients. The transduced cells exhibited multifunctional responses against mutated peptides and maintained long-term activity within patients. These findings highlight the safety, feasibility, and efficacy of TCR therapy, providing a personalized immunotherapeutic strategy for patients with non-MSI-H CRC[23].

CHALLENGES OF CAR-T IN CRC

Despite the promise of ACT, particularly CAR-T cell therapy, several challenges remain in the clinical application of CRC treatment (Figure 4).

Figure 4 Challenges in adoptive cell therapy for colorectal cancer. Schematic diagram of the challenges in adoptive cell therapy for colorectal cancer, including immune evasion, immunosuppressive tumor microenvironment, insufficient immune cell infiltration, antigen heterogeneity, off-target effects, cytokine release syndrome, and immune effector cell-associated neurotoxicity syndrome. TME: Tumor microenvironment; CRS: Cytokine release syndrome; ICANS: Immune effector cell-associated neurotoxicity syndrome. (Created with BioRender.com).

Antigen heterogeneity and immune evasion

CRC exhibits pronounced antigen heterogeneity, both between patients and within individual tumors. The expression levels of common target antigens, such as EpCAM, CEA, mucin 1 (MUC1), receptor tyrosine kinase-like orphan receptor 1 (ROR1), GPA33, and c-Met are highly variable, limiting the efficacy of single-target ACT, including CAR-T and TCR-T approaches[24].

Under immune pressure from ACT, tumor cells can undergo immunoediting and evolve to escape immune surveillance. This includes downregulation or complete loss of antigen expression, as well as HLA class I downregulation, both of which impair T-cell recognition[25]. Antigen loss is a well-documented mechanism of acquired resistance, contributing to relapse in ACT-treated malignancies. For instance, CD19-negative relapse has been frequently observed in B-acute lymphoblastic leukemia, while loss of B-cell maturation antigen and EGFR VIII has been reported in multiple myeloma and glioblastoma, respectively. Similar escape phenomena may occur in CRC, particularly given its high intratumoral heterogeneity[26].

These challenges highlight the necessity for multi-antigen targeting strategies, optimized antigen selection, and the integration of combination therapies, such as immune checkpoint inhibitors, to enhance antigen presentation and reduce immune escape[27]. Advances in neoantigen prediction and single-cell profiling may also help tailor ACT to the antigenic landscape of individual tumors[28].

Immunosuppressive tumor microenvironment

The CRC tumor microenvironment significantly impairs antitumor immune responses by secreting immunosuppressive factors, such as IDO1, programmed cell death ligand 1 (PD-L1), and IL-10, or by recruiting immunosuppressive cells. These cells include tumor-associated macrophages and myeloid-derived suppressor cells[29]. Additionally, its hypoxic microenvironment also contributes to immune suppression. The hypoxic environment can induce CD39 expression, enhancing the formation of adenosine in the immune microenvironment and inhibiting immune cell function[30]. Meanwhile, it stimulates mitochondrial stress in T cells, accelerating the T cell exhaustion, thereby restricting anti-tumor immunity[31,32]. This complex immunosuppressive environment significantly limits the efficacy of ACT[33].

Insufficient immune cell infiltration

The physical barriers of solid tumors can restrict adoptive immune cells from reaching the tumor core[34]. For example, elevated stromal hydraulic pressure, together with a dense extracellular matrix block efficient interaction of adoptive immune cells with tumor cells. Moreover, the physical blockage can result in early exhaustion of adoptive immune cells with low cytotoxic activities[35]. In addition, the lack of immune-promoted chemokines in the tumor microenvironment further hampers infiltration, diminishing therapeutic efficacy[36].

Off-target effects

CRC typically lacks truly TSAs and instead expresses tumor-associated antigens (TAAs)[37]. CAR-T and TCR-T cells targeting TAAs may induce off-target effects because antigen may also be expressed in normal tissues, thus resulting in damage to healthy cells and severe side effects[38]. The low mutation frequency of TSAs in CRC further limits the development of TSA-targeted therapies. Current research focuses on optimizing CAR and TCR designs to minimize off-target effects[39]. Off-target toxicity is rarely observed in TIL-based therapies, which may be attributed to the natural negative selection of self-reactive TCRs during thymic development. This process reduces the likelihood of TILs recognizing and attacking normal tissues, contributing to the favorable safety profile of TIL therapy[40].

Cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome

Cytokine release syndrome (CRS) is a systemic inflammatory response that can occur following ACT, including CAR-T, TCR-T, and TIL treatments. This response is caused by immune cell overactivation, resulting in massive cytokine release, particularly IL-6, IFN-γ, and tumor necrosis factor-α[41]. While CRS is more frequently observed with CAR-T therapy, it can also occur with TCR-T and TIL therapies, though with less severity. Tumor cell pyroptosis and subsequent macrophage activation amplify the cytokine cascade, resulting in endothelial dysfunction, vascular leakage, and organ toxicity[42]. A rapid surge in cytokine levels can result in symptoms such as fever, hypotension, and respiratory distress. In severe cases, this may progress to multi-organ dysfunction[43].

Immune effector cell-associated neurotoxicity syndrome (ICANS) is primarily driven by blood-brain barrier disruption due to elevated systemic cytokines following CAR T-cell therapy[44]. This allows immune cells and cytokines to infiltrate the central nervous system, activating microglia and astrocytes. The resulting neuroinflammatory cascade contributes to cerebral edema, neuronal dysfunction, and a range of neuropsychiatric symptoms, including confusion, expressive aphasia, and encephalopathy[45]. Although rare, ICANS is a serious and potentially life-threatening toxic syndrome. Effective management of these toxicities, such as using tocilizumab to inhibit IL-6 receptors, is essential for the clinical use of CAR-T therapy[25].

STRATEGIES TO OVERCOME CHALLENGES

Modulating the tumor microenvironment

The immunosuppressive properties of the tumor microenvironment are key factors that limit ACT efficacy. Modulating immune cell balance within the tumor microenvironment can greatly enhance antitumor activity.

B7-H3 is an immunosuppressive molecule highly expressed in various tumors that is closely linked to immune evasion. Research indicates that B7-H3 overexpression activates the signal transducers and activators of transcription (STAT3) signaling pathway, which reduces the expression of the natural killer group 2 member D (NKG2D) ligand ULBP2[46]. This impairs T-cell-mediated tumor recognition and cytotoxicity. Knockdown of B7-H3 or inhibition of STAT3 phosphorylation restores ULBP2 expression, significantly enhancing the antitumor efficacy of ACT. A negative correlation has been reported between B7-H3 and ULBP2 expression in CRC[46]. This implies that targeting the B7-H3 or STAT3/ULBP2 axis is be a potential therapeutic strategy.

Proinflammatory cytokines, such as IL-7, IL-12, and IL-15, have shown potential in countering tumor microenvironment immunosuppression, but they may cause systemic toxicity[47,48]. Studies suggest that genetic engineering to introduce IL-7 and the chemokine XCL1 enhances intratumoral type I conventional dendritic cell recruitment[49]. This promotes endogenous CD8 + T cell responses to tumor neoantigens[49]. Similarly, CRISPR/Cas9-mediated regulation of CXCR2 and IL-2 expression in natural killer-92 cells significantly enhances their antitumor efficacy in CRC[50].

T cells redirected for universal cytokine-mediated killing (TRUCKs) are fourth-generation CAR-T cells designed for improved treatment of solid tumors. They are engineered to integrate inducible cytokines like nuclear factor of activated T-cells to release immunomodulatory cytokines upon CAR activation, including IL-12, IL-18, and IL-15, [51]. These cytokines act locally within the tumor microenvironment to enhance immune cell infiltration, activate natural killer cells and macrophages, and promote memory T-cell formation[52]. This converts immunologically “cold” tumors into “hot” ones. As a result, TRUCKs improve antitumor activity and T-cell persistence[53]. TRUCKs reduce systemic toxicity compared to traditional CAR-T cells. Ongoing clinical trials are testing various cytokine combinations to optimize therapeutic outcomes. Additionally, CAR-T cells that target transforming growth factor-β (TGF-β) demonstrated significantly improved antitumor activity[54,55]. Other studies have shown that engineered mesenchymal stem cells can deliver CXCL9 and OX40L to remodel the tumor microenvironment, promoting effector CD8 + T cell infiltration and activation, offering innovative combination therapies for CRC[56].

Enhancing T-cell activity and survival

The efficacy of ACT largely depends on T-cell function and their adaptability within the solid tumor microenvironment[57]. Advances in gene editing and molecular engineering have significantly enhanced T-cell activity and persistence[58]. Nuclear factor kappa-B-inducing kinase, an important regulator of T-cell mitochondrial metabolism and oxidative phosphorylation, can improve T-cell metabolic fitness and antitumor efficacy[59].

To address T-cell exhaustion, CRISPR-Cas9-mediated deletion of the Cbl-b gene effectively inhibits T-cell functional decline, significantly improving CAR-T cell persistence and antitumor efficacy in the tumor microenvironment[60]. Furthermore, the immunosuppressive molecule mitochondrial arginase-2 (Arg2) is a target to optimize CD8 + T-cell activation and function. Preclinical studies suggest that knocking out Arg2 enhances CD8 + T-cell persistence and effectively inhibits tumor growth in CRC models[61].

Optimizing CAR-T design

Co-stimulatory signaling domains (e.g., CD28 and 4-1BB) are crucial for CAR-T cell proliferation and antitumor efficacy[62]. For example, introducing CD27 co-stimulatory signaling into CAR-T cells resulted in CEA28BB27Z CAR-T cells, which had enhanced proliferation, antitumor activity, and persistence in colorectal tumor treatment[63]. These improvements were linked to reduced immune checkpoint receptor expression, increased generation of CD4 + and CD8 + stem cell-like memory T cells, and a more stable fused mitochondrial network[63]. Additionally, incorporating CD8α structural domains into CAR-T cells targeting GUCY2C enhanced immunological synapse formation, promoted T-cell activation, and increased effector molecule secretion, significantly improving antitumor efficacy[64].

Dual-targeting and novel therapeutic approaches

Dual-targeting CAR-T cell strategies are broadly categorized into two main approaches. The first involves the design of tandem or bispecific CAR constructs capable of simultaneously recognizing two distinct TAAs, thereby enhancing tumor specificity and reducing the likelihood of antigen escape. The second employs logic-gated systems, such as “AND: or “OR” gate configurations, where CAR-T cell activation is conditioned upon the co-recognition of multiple antigenic signals. This design improves targeting precision and minimizes off-tumor cytotoxicity.

Building on these innovative strategies, dual-targeting CAR-T cell designs through genetic engineering have emerged as a promising strategy to address tumor heterogeneity and antigen loss[65]. A nonrandomized clinical trial evaluated CAR-T cells targeting both CD19 and guanylate cyclase-C (GCC) in patients with metastatic CRC. The results showed a significant reduction in tumor burden in some patients and good treatment tolerability[66]. These findings suggest that dual-antigen-targeting CAR-T cells have significant potential for treating metastatic CRC, particularly in patients who are unresponsive to standard therapies.

A novel therapeutic approach combining human epidermal growth factor receptor 2 (HER2)-synNotch and CEA-CAR technologies has been developed. This strategy involves engineering nature killer-92 cells to express a synNotch receptor targeting HER2, which activates CEA-CAR expression upon HER2 signaling. The system demonstrated significant antitumor activity against HER2-amplified CRC both in vitro and in vivo. Importantly, there was no toxicity to normal tissues with physiological HER2 levels. Although based on a CAR-nature killer platform, this strategy is designed to overcome key limitations frequently encountered in conventional CAR-T cell therapies, particularly off-tumor toxicity. By incorporating a logic-gated mechanism, this approach offers a safe, effective, and scalable treatment for HER2-amplified CRC[67].

Transient expression

To enhance safety and controllability, researchers have developed electroporation-based methods to optimize CAR-T cell therapy through transient RNA-CAR expression[68]. This non-permanent genetic modification allows T cells to temporarily express new CARs, enabling them to target specific antigens on cancer cells. This approach has shown enhanced antitumor activity in solid tumors like CRC, while reducing long-term side effects, lowering toxicity risks, and avoiding issues associated with viral vectors, including CRS and off-target effects[69]. Additionally, the therapeutic effects are dose-dependent[69]. This technology has also been effectively used in CAR-nature killer cells expressing CXCR1 and NKG2D, improving their migration and tumor infiltration capabilities, offering a novel treatment approach for refractory cancers[70]. Furthermore, with their short lifespan and immune tolerance, CAR-nature killer cells pose lower toxicity risks compared to CAR-T cells in CRC treatment. This makes them a safer and effective alternative for adoptive immunotherapy in CRC[71]. Future research should focus on optimizing their in vivo persistence and tumor infiltration capacity to further enhance therapeutic efficacy.

Enhancing tumor immune cell infiltration

The physical barriers of the solid tumor microenvironment, such as a dense extracellular matrix and elevated stromal hydraulic pressure, limit ACT tumor infiltration. Combining adoptive immune cells with matrix-degrading enzymes like hyaluronidase can improve tumor stroma permeability and enhance CAR-T cell penetration[72]. Another promising strategy involves vascular disrupting agents[73]. Studies have shown that CA4P promotes IFN-γ production by CAR-T cells, significantly enhances CAR-T cell infiltration in solid tumors, and improves therapeutic outcomes. In preclinical studies of murine models of colon and ovarian cancers, as well as patient-derived xenograft models, the combination of CA4P with CAR-T therapy significantly boosted antitumor efficacy. These findings suggest that CA4P could serve as an adjunctive agent to overcome delivery barriers in CAR-T therapy for solid tumors, offering a valuable approach for CRC immunotherapy[73].

Combination therapies

Combining CAR-T cells with other approaches is a key strategy to enhance efficacy and overcome immune evasion. For example, surgery, radiochemotherapy, immune checkpoint inhibitors (e.g., PD-1/PD-L1 inhibitors), and oncolytic virus therapy can reduce tumor burden, promote antigen release, and improve tumor stroma permeability[74-76]. Checkpoint inhibitors, such as anti-PD-1 and anti-cytotoxic T-lymphocyte-associated protein 4 therapies[77], effectively restore T-cell function. They also reduce the immunosuppressive effects of the tumor microenvironment[78]. PD-1 is an inhibitory receptor on T cells. It binds to its ligand PD-L1, causing T-cell exhaustion and loss of antitumor function. Gene-editing technologies like CRISPR/Cas9 can modify the PD-1 gene[79] to alleviate tumor microenvironment-mediated immunosuppression and improve CAR-T cell effectiveness in preclinical models[80]. Additionally, PD-1 blocking agents can continuously inhibit the PD-1/PD-L1 signaling axis. This restores CAR-T cell activity and enhances tumor cytotoxicity[81]. Combining other immune checkpoint inhibitors, such as LAG3 inhibitors, further strengthens antitumor immune responses[82]. These effects create a more favorable tumor microenvironment, facilitating immune cell infiltration and enhancing antitumor responses[36,83]. Additionally, research has demonstrated that the effects of T CD8 + cells persist after combined chemotherapy drug treatment, which is important for the study of combined chemotherapy and CAR-T cell therapy[84]. However, further clinical studies are necessary to substantiate the clinical application potential of combined therapy in CRC.

Innovative delivery technologies

Optimizing cell delivery routes is an important area of research. Studies have shown that intraperitoneal injection enhances CAR-T cell antitumor activity and persistence in peritoneal carcinoma models compared to intravenous administration[85]. This offers a novel delivery strategy for solid tumors like CRC. The HITM-SIR phase Ib clinical trial further validated the safety and bioactivity of intrahepatic arterial (HAI) administration of anti-CEA CAR-T cells in combination with selective internal radiation therapy in patients with CEA-positive liver metastases. Six patients with multiple systemic therapy failure were included. The results showed that CAR-T HAI infusion did not cause severe toxicity or CRS. Treatment reduced inflammatory cytokine levels, stabilized or decreased serum CEA levels, and achieved an average survival time of 11 months[86]. This trial provides the first evidence supporting localized adoptive immunotherapy to reduce systemic toxicity and enhance antitumor activity, offering a new direction for treating CRC liver metastases[86].

Research on novel targets

Current CAR-T cell therapies primarily focus on TAAs to enhance efficacy while minimizing side effects. In CRC, potential targets under investigation include EpCAM[87,88], CEA[86], MUC1, EGFR[89], ROR1 (an onco-embryonic antigen)[90], placental alkaline phosphatase, CD318[91], GPA33[92], and c-Met[93]. These targets are of particular interest due to their high expression in CRC cells, making them crucial areas of CAR-T therapy research.

Glycosylation is a key post-translational modification that plays an important role in CRC development and progression[94]. Aberrant glycosylation patterns, such as Tn and STn antigens, form specific tumor epitopes that serve as precise targets for CAR-T cell therapy. These epitopes can significantly enhance tumor-killing efficacy while reducing off-target risks[95]. Cancer stem cells (CSCs), a critical subpopulation of tumor cells that drive CRC tumor progression and resistance, are also emerging as key targets. These cells express specific markers such as CD133 and ALDH1. However, the targeting efficacy of current ACT approaches to CSCs is limited by immune evasion mechanisms, including high PD-L1 expression and immunosuppressive factor secretion[96]. Future strategies that combine CSC marker and glycosylation epitope targeting with immune checkpoint inhibitors and CSC signaling pathway inhibitors (e.g., Wnt and Notch pathways) show promise in overcoming these therapeutic challenges. These approaches could significantly improve CRC immunotherapy efficacy and safety. This article summarizes recent clinical studies targeting different CAR-T cell therapy targets for CRC (Table 3).

Table 3 Clinical trials of chimeric antigen receptor-T cell therapy for colorectal cancer.

Antigen	NCT number	Sponsor	Phases	Start date	Completion date
CEA	NCT04513431	Ruijin Hospital	EARLY_PHASE1	August 30, 2020	August 30, 2023
	NCT02349724	Southwest Hospital, China	PHASE1	December 1, 2014	December 1, 2019
	NCT02959151	Shanghai GeneChem Co., Ltd.	PHASE1|PHASE2	July 1, 2016	July 1, 2018
	NCT05396300	Fang WJ, MD	PHASE1	May 25, 2022	May 15, 2025
	NCT05736731	A2 Biotherapeutics Inc.	PHASE1|PHASE2	April 28, 2023	December 1, 2028
	NCT05240950	Changhai Hospital	PHASE1	August 25, 2022	December 25, 2025
	NCT06043466	Chongqing Precision Biotech Co., Ltd	PHASE1	August 11, 2023	December 31, 2027
	NCT03682744	Sorrento Therapeutics, Inc.	PHASE1	September 13, 2018	March 1, 2021
	NCT02850536	Roger Williams Medical Center	PHASE1	February 1, 2017	September 17, 2021
GUCY2C	NCT06718738	Beijing Immunochina Medical Science and Technology Co., Ltd.	PHASE1	December 15, 2024	August 15, 2026
	NCT06675513	Wondercel Biotech (ShenZhen)	EARLY_PHASE1	November 10, 2024	December 30, 2029
	NCT05319314	Innovative Cellular Therapeutics Inc.	PHASE1	August 1, 2022	October 1, 2024
MSLN	NCT05089266	Shanghai Cell Therapy Group Co., Ltd	PHASE1	November 30, 2021	January 30, 2025
	NCT06051695	A2 Biotherapeutics Inc.	PHASE1|PHASE2	April 3, 2024	June 1, 2029
	NCT06256055	UTC Therapeutics Inc.	PHASE1	March 5, 2024	April 1, 2025
NKG2DL	NCT05248048	The Third Affiliated Hospital of Guangzhou Medical University	EARLY_PHASE1	September 13, 2021	October 1, 2022
	NCT04550663	The Affiliated Nanjing Drum Tower Hospital of Nanjing University Medical School	PHASE1	September 25, 2020	March 25, 2023
	NCT04107142	CytoMed Therapeutics Pte Ltd	PHASE1	December 1, 2019	March 1, 2021
	NCT04991948	Celyad Oncology	PHASE1	November 22, 2021	May 25, 2038
MUC1	NCT02617134	PersonGen BioTherapeutics (Suzhou) Co., Ltd.	PHASE1|PHASE2	November 1, 2015	November 1, 2018
	NCT05239143	Poseida Therapeutics, Inc.	PHASE1	February 15, 2022	April 1, 2039
CDH17	NCT06055439	Chimeric Therapeutics	PHASE1|PHASE2	May 15, 2024	May 1, 2027
C-met	NCT03638206	Shenzhen BinDeBio Ltd.	PHASE1|PHASE2	March 1, 2018	March 1, 2023
EGFR	NCT06682793	A2 Biotherapeutics Inc.	PHASE1|PHASE2	March 31, 2025	March 31, 2030
EpCAM	NCT05028933	Zhejiang University	PHASE1	September 30, 2021	December 31, 2024
HER2	NCT02713984	Yang Z	PHASE1|PHASE2	March 1, 2016	July 1, 2019
	NCT03740256	Baylor College of Medicine	PHASE1	December 14, 2020	December 30, 2038
CD133	NCT02541370	Chinese PLA General Hospital	PHASE1|PHASE2	June 1, 2015	June 1, 2019
LGR5	NCT05759728	Carina Biotech Limited	PHASE1|PHASE2	October 24, 2023	December 1, 2027
B7-H3	NCT05190185	PersonGen BioTherapeutics (Suzhou, Jiangsu Province) Co., Ltd.	PHASE1	June 1, 2021	December 1, 2023

CEA: Carcinoembryonic antigen; NKG2DL: Natural killer group 2D receptor; MUC1: Mucin 1; C-met: Cellular-mesenchymal epithelial transition; EGFR: Epidermal growth factor receptor; EpCAM: Epithelial cell adhesion molecule; HER2: Human epidermal growth factor receptor 2; CD: Cluster of differentiation; NCT: National clinical trials.

CONCLUSION

ACT has shown strong potential in treating CRC. Traditional TIL therapy involves isolating and expanding TILs from the patient’s tumor, but its clinical application is limited by technical complexity and high costs. TCR therapy enhances T-cell specificity for TAAs through genetic engineering of TCRs, but it remains susceptible to MHC downregulation and off-target effects. In contrast, CAR-T therapy bypasses MHC dependency by using single-chain variable fragments to activate antitumor immune responses. However, the application of CAR-T therapy in solid tumors like CRC is limited by antigen heterogeneity, the immunosuppressive tumor microenvironment, and off-target toxicity. As an alternative, CAR-nature killer therapy presents potential as an “off-the-shelf” immunotherapy due to its reduced risk of CRS and the safety of allogeneic transplantation. Studies show that CAR-nature killer cells exhibit advantages in tumor infiltration and cytotoxicity while causing minimal damage to normal tissues. To facilitate clinical application, patient selection criteria for ACT in CRC are increasingly being defined by tumor molecular subtype, antigen expression profiles (e.g., CEA, HER2, GCC), microsatellite stable (MSS)/MSI, and immune landscape (hot vs cold tumors). Identifying patients with actionable antigen targets and favorable immunogenicity remains critical for ACT efficacy. Several ongoing early-phase clinical trials are exploring next-generation CAR-T therapies for solid tumors, including CRC. These include logic-gated CAR-T cells targeting CEA or MSLN in the context of HLA-A 02 loss (NCT05736731, NCT06051695), and CAR-T strategies combined with oncolytic viruses or immune checkpoint inhibitors (NCT05089266, NCT03740256). While most remain in phase I, these trials reflect a growing emphasis on improving antigen specificity and enhancing CAR-T therapy efficacy for solid tumors. Regulatory challenges, including issues with manufacturing consistency, scalability, and the management of off-target effects, continue to pose significant barriers to Food and Drug Administration approval. Overcoming these translational hurdles is crucial for the widespread adoption of CAR-T therapy in CRC. In personalized treatment, integrating genomic, transcriptomic, and immunomic data to identify neoantigens or TAAs enables tailored therapeutic regimens. This approach significantly improves efficacy while minimizing side effects. Neoantigen-based therapies have shown promise in MSS CRC and immune “cold” tumors. However, challenges such as low mutational burden and immune evasion mechanisms remain. Future directions should focus on the rational optimization of CAR structures to enhance antigen sensitivity and reduce T-cell exhaustion. Additionally, the development of multi-specific CARs or logic-gated designs is crucial to overcome antigen heterogeneity. Strategies aimed at prolonging in vivo T-cell persistence, such as cytokine support, memory phenotype induction, or gene editing, should also be prioritized. Improving delivery methods, including regional perfusion and focused injection techniques, may enhance tumor infiltration. Furthermore, combining CAR-T cells with checkpoint inhibitors, TGF-β blockers, or oncolytic viruses could help reshape the tumor microenvironment. These approaches collectively aim to improve ACT safety and efficacy, expanding its clinical potential in solid tumors. With ongoing advancements in cell engineering, manufacturing, and delivery technologies, CAR-T holds significant potential to become a transformative component of CRC immunotherapy. Future research should focus on integrating CAR-T with current treatment paradigms, such as chemotherapy, targeted therapies, and immune checkpoint blockade, to achieve more durable responses. Moreover, real-world validation through multicenter clinical trials and long-term follow-up studies will be crucial for successful clinical translation.