Abstract
Gastrointestinal (GI) malignancies account for more than one-third of cancer-related deaths worldwide, and outcomes for advanced disease remain poor despite advances in systemic therapy and immunotherapy. Chimeric antigen receptor (CAR)-engineered cellular therapies have transformed the treatment of hematologic malignancies, but their application in solid tumors has been limited by antigen heterogeneity, immunosuppressive tumor microenvironments, on-target/off-tumor toxicity, and manufacturing challenges. This narrative review examines the clinical evidence supporting chimeric antigen receptor T-cell (CAR-T) and chimeric antigen receptor natural killer (CAR-NK) therapies directed against six major targets in GI oncology: claudin 18.2 (CLDN18.2), guanylyl cyclase C (GUCY2C), mesothelin (MSLN), glypican-3 (GPC3), carcinoembryonic antigen (CEA), and human epidermal growth factor receptor 2 (HER2). Among these, CLDN18.2 is the most advanced target; in the randomized phase II CT041-ST-01 trial, satricabtagene autoleucel significantly improved progression-free survival (PFS) compared with physician’s choice therapy in previously treated gastric or gastroesophageal junction cancer, providing the first randomized evidence of CAR-based therapy efficacy in a solid tumor. GUCY2C and GPC3 have shown promising early-phase activity in metastatic colorectal cancer and hepatocellular carcinoma, respectively, whereas mesothelin, CEA, and HER2 continue to face important biological and safety challenges. CAR-NK cells have emerged as a complementary platform with favorable safety characteristics and off-the-shelf manufacturing potential, although clinical experience in GI malignancies remains limited. We discuss the relative clinical maturity of these targets, key barriers to implementation, and emerging strategies that may advance the role of CAR-based cellular therapies in GI cancers.
Keywords: adoptive cell therapy, CAR-NK cells, CAR-T cell therapy, claudin 18.2, colorectal cancer, gastric cancer, gastrointestinal cancers, glypican-3
1. Introduction
Gastrointestinal (GI) malignancies, including cancers of the esophagus, stomach, colorectum, liver, pancreas, and biliary tract, represent a major global health burden. In 2022, GI cancers accounted for approximately 4.9 million new cases and 3.3 million deaths worldwide, representing more than one-third of all cancer-related mortality (1). The global burden is projected to nearly double by 2050, with the greatest increases expected in low- and middle-income countries (1).
Despite substantial advances in surgery, systemic therapy, targeted agents, and immunotherapy (IO), outcomes for patients with advanced GI cancers remain poor (2–11). Median overall survival for metastatic pancreatic ductal adenocarcinoma remains approximately 9–11 months with current first-line multi-agent chemotherapy, while first-line chemotherapy for advanced gastric cancer yields a median overall survival of less than 12 months in the absence of IO (12, 13). Immune checkpoint inhibitors (ICIs) have transformed treatment for selected molecular subgroups. In microsatellite instability-high (MSI-H) or mismatch repair-deficient (dMMR) colorectal cancer, nivolumab plus ipilimumab demonstrated superior progression-free survival (PFS) compared with chemotherapy in the CheckMate 8HW trial (14). Similarly, pembrolizumab-based regimens have produced durable responses in MSI-H gastric and gastroesophageal cancers, with objective response rates (ORR) ranging from 46% to 65% (15). In the first-line setting, nivolumab combined with chemotherapy improved overall survival (OS) in programmed death-ligand 1 (PD-L1) combined positive score (CPS) ≥5 gastric and gastroesophageal junction (GEJ) adenocarcinoma in the CheckMate 649 trial (median OS 14.4 vs. 11.1 months; hazard ratio (HR) 0.71; p < 0.0001) (12). Pembrolizumab plus chemotherapy demonstrated similar benefit across PD-L1 subgroups in KEYNOTE-859 (median OS 12.9 vs. 11.5 months in the intention-to-treat (ITT) population; HR 0.78; p < 0.0001) (16). However, most GI tumors, including pancreatic ductal adenocarcinoma (PDAC), most hepatocellular carcinomas (HCC), and microsatellite-stable colorectal cancers (CRC), exhibit primary or acquired resistance to checkpoint blockade (8, 11, 17–23). As a result, durable disease control remains difficult to achieve for most patients (8, 11, 17–23).
Adoptive cellular IO has emerged as one of the most important advances in modern oncology. Among these approaches, chimeric antigen receptor (CAR)-engineered immune cells have achieved remarkable success in hematologic malignancies. CD19-directed chimeric antigen receptor T-cell (CAR-T) therapies have produced durable remissions in relapsed or refractory B-cell acute lymphoblastic leukemia and large B-cell lymphoma, while B-cell maturation antigen (BCMA)-targeted CAR-T cells have significantly improved outcomes in multiple myeloma (24, 25). Multiple CAR-T products have now received regulatory approval for hematologic cancers, establishing cellular therapy as a new treatment pillar alongside surgery, radiation, systemic therapy, and systemic pharmacotherapy (24).
These achievements have generated considerable interest in extending CAR-based therapies to solid tumors, including GI malignancies. However, translating CAR therapy from hematologic cancers to solid tumors has proven far more challenging. GI tumors are characterized by marked antigen heterogeneity, dense desmoplastic stroma, abnormal tumor vasculature, impaired immune-cell trafficking, and profoundly immunosuppressive tumor microenvironments (TME) (26, 27). In addition, most candidate targets are tumor-associated rather than truly tumor-specific, creating a risk of on-target, off-tumor toxicity due to low-level expression in normal tissues (28). Antigen escape, T-cell exhaustion, and limited in vivo persistence further restrict therapeutic efficacy. Consequently, CAR-based therapies have not yet achieved the level of success in solid tumors that has been observed in hematologic malignancies (24).
Despite these challenges, the field has advanced rapidly over the past decade. Improvements in CAR engineering, including optimized costimulatory domains, armored CAR constructs, cytokine-secreting platforms, logic-gated systems, affinity-tuned receptors, and safety-switch technologies, have expanded the therapeutic potential of cellular therapy in solid tumors (26). At the same time, advances in tumor biology have led to the identification of several clinically relevant targets across GI malignancies, including claudin 18.2 (CLDN18.2), guanylyl cyclase C (GUCY2C), mesothelin (MSLN), glypican-3 (GPC3), carcinoembryonic antigen (CEA), and human epidermal growth factor receptor 2 (HER2). Notably, CLDN18.2-directed CAR-T therapy (satricabtagene autoleucel) recently produced the first positive randomized controlled trial result for a CAR-based therapy in a solid tumor, demonstrating a significant progression-free survival benefit over physician’s choice treatment in previously treated gastric cancer (29). Although no CAR-based therapy has yet received regulatory approval for a solid tumor indication, these findings provide compelling proof-of-concept that engineered cellular therapies can overcome at least some of the barriers that have historically limited efficacy in GI cancers.
Alongside CAR-T cells, CAR-engineered natural killer (CAR-NK) cells have emerged as a complementary platform with several potential advantages. In the landmark study by Liu et al. (2020), cord blood-derived anti-CD19 CAR-NK cells armed with interleukin-15 achieved a 73% response rate (8 of 11 patients) in a phase I/II study of patients with relapsed or refractory B-cell malignancies without causing cytokine release syndrome, neurotoxicity, or graft-versus-host disease (GvHD) (30). CAR-NK cells retain endogenous receptor-mediated cytotoxicity, exhibit a favorable safety profile, and can be manufactured as allogeneic off-the-shelf products from cord blood, peripheral blood, or induced pluripotent stem cells (31, 32). These characteristics make CAR-NK therapies particularly attractive for solid tumors, where scalable manufacturing and repeated dosing may be required.
Given the growing body of clinical evidence, a focused evaluation of the most advanced CAR-based targets in GI oncology is warranted. In this review, we examine the biological rationale, clinical development, efficacy, and safety of CAR-T and CAR-NK therapies targeting CLDN18.2, GUCY2C, mesothelin, GPC3, CEA, and HER2. We also discuss key translational challenges, compare outcomes across targets and platforms, and highlight future directions that may shape the role of cellular IO in GI cancers.
Given the limited clinical maturity of CAR-NK therapy in GI malignancies, CAR-NK evidence is discussed primarily to provide biological and translational context rather than as a systematic assessment of preclinical CAR-NK targets.
2. CAR-T and CAR-NK cell therapy: biological principles
2.1. Evolution and structure of chimeric antigen receptors
Chimeric antigen receptors (CARs) are synthetic receptors that redirect immune cells to recognize tumor-associated surface antigens independently of major histocompatibility complex (MHC) presentation. A conventional CAR consists of an extracellular antigen-binding domain, most commonly a single-chain variable fragment (scFv), linked through a hinge and transmembrane region to intracellular signaling domains that activate the effector cell upon antigen engagement (33, 34) (Figure 1).
Figure 1.

Structure and evolution of chimeric antigen receptors (CARs). (A) Modular architecture of a conventional CAR consisting of an extracellular single-chain variable fragment (scFv) antigen-binding domain, hinge/spacer region, transmembrane domain, and intracellular signaling domains. (B) Evolution of CAR designs from first-generation constructs containing CD3ζ alone to advanced engineered platforms incorporating cytokine-secreting and cytokine-signaling elements. All currently approved CAR-T products are based on second-generation designs incorporating either CD28 or 4-1BB costimulation. Created with BioRender.com.
Although multiple generations of CARs have been developed, nearly all clinically successful CAR-T products currently use second-generation constructs incorporating a single costimulatory domain, typically CD28 or 4-1BB, in addition to CD3ζ signaling (35). The choice of costimulatory domain influences expansion kinetics, persistence, metabolic fitness, and susceptibility to exhaustion, making CAR design directly relevant to clinical outcomes. Third-generation CARs incorporate two costimulatory domains (e.g., CD28 and 4-1BB) to enhance signaling, although clinical superiority over second-generation designs has not been consistently demonstrated. Fourth-generation CARs, also termed TRUCKs (T cells redirected for universal cytokine killing), are engineered to secrete transgenic cytokines or other effector molecules upon CAR activation (36). More recently, fifth-generation constructs have incorporated intracellular domains from cytokine receptors (e.g., IL-2Rβ) to activate JAK-STAT signaling, further enhancing proliferation and persistence (37).
More advanced platforms have been developed to address the challenges of solid tumors. These include armored CARs that secrete cytokines or resist immunosuppressive signaling, logic-gated systems that improve tumor specificity, and next-generation constructs designed to enhance persistence and function within hostile TMEs (36, 37). Several of these approaches are now being incorporated into clinical products targeting GI malignancies and will be discussed throughout this review.
2.2. CAR-T cell therapy
CAR-T cell therapy involves the genetic modification of autologous T lymphocytes to express CARs directed against tumor-associated antigens. Following ex vivo expansion and reinfusion, CAR-T cells recognize target antigens independently of MHC presentation and eliminate tumor cells through perforin-granzyme release, Fas/FasL-mediated apoptosis, cytokine secretion, and recruitment of endogenous immune responses. The clinical success of CAR-T therapy was established in hematologic malignancies, with tisagenlecleucel achieving an 81% remission rate in pediatric and young adult patients with relapsed or refractory B-cell acute lymphoblastic leukemia (38), and subsequent approvals in lymphoma and multiple myeloma confirming the ability of CAR-T cells to induce durable responses in treatment-refractory disease (39, 40).
Applying this approach to GI solid tumors has been far more difficult. The hostile TME, poor antigen uniformity, and physical stromal barriers that characterize these cancers have collectively blunted CAR-T efficacy and motivated many of the engineering strategies now entering clinical testing.
2.3. CAR-NK cell therapy
Natural killer (NK) cells are innate lymphoid cells capable of recognizing and eliminating malignant cells without prior antigen sensitization. NK-cell activation is governed by a balance between activating and inhibitory receptor signaling. Inhibitory receptors, including killer-cell immunoglobulin-like receptors (KIRs) and CD94/NKG2A, recognize MHC class I molecules on healthy cells and suppress cytotoxic activity. Downregulation of MHC class I expression, a common mechanism of tumor immune evasion, removes inhibitory signaling and promotes NK-cell activation through “missing-self” recognition (41).
Activation is further driven by receptors such as NKp46, NKp30, NKp44, NKG2D, and CD16. NKG2D recognizes stress-induced ligands expressed on transformed cells, whereas CD16 mediates antibody-dependent cellular cytotoxicity (ADCC) (42). Activated NK cells kill tumor cells through perforin-granzyme release and death receptor-mediated apoptosis involving Fas ligand (FasL) and TRAIL pathways (43).
CAR-NK cells combine this intrinsic cytotoxic machinery with engineered antigen specificity. Importantly, CAR-NK cells retain their native killing mechanisms, allowing both CAR-dependent and CAR-independent tumor elimination, a feature that may be particularly advantageous in solid tumors where antigen heterogeneity and antigen loss frequently contribute to therapeutic resistance. Several cellular sources have been explored for CAR-NK manufacturing, including peripheral blood, umbilical cord blood, induced pluripotent stem cells (iPSCs), and NK-cell lines such as NK-92, facilitating the development of standardized allogeneic off-the-shelf products.
Clinical feasibility was established by Liu et al. (2020), who infused cord blood-derived anti-CD19 CAR-NK cells co-expressing IL-15 and an inducible caspase-9 safety switch into patients with relapsed or refractory CD19-positive malignancies (30). Responses occurred in 73% of treated patients, including seven complete remissions, without cytokine release syndrome (CRS), neurotoxicity, or GvHD. Subsequent follow-up confirmed the favorable safety profile and sustained clinical activity (44). More recently, Lei et al. (2025) reported a 62.5% response rate, including 50% complete remissions, using 4-1BB-costimulated CAR-NK cells in relapsed or refractory large B-cell lymphoma, again without CRS, neurotoxicity, or GvHD (45). Importantly, despite encouraging clinical experience in hematologic malignancies, clinical evidence for CAR-NK therapy in GI malignancies remains limited and substantially less mature than that for CAR-T therapy. Wang et al. (2025) reported results of NKG2D-directed CAR-NK cells combined with or without PD-1 blockade in metastatic colorectal cancer (45), while early-phase GPC3-directed CAR-NK studies in HCC are ongoing but have not yet reported mature efficacy data.
2.4. CAR-T versus CAR-NK cells: comparative biology
Although both CAR-T and CAR-NK therapies redirect immune cells toward tumor-associated antigens, important biological and clinical differences distinguish the two platforms. CAR-T cells exhibit robust antigen-driven expansion, prolonged persistence, and the capacity to generate memory-like populations capable of long-term immune surveillance. These characteristics contribute to the durable responses observed in hematologic malignancies but also increase the risk of treatment-related toxicities, including CRS, immune effector cell-associated neurotoxicity syndrome (ICANS), prolonged cytopenias, and on-target/off-tumor toxicity. Autologous CAR-T manufacturing remains complex, expensive, and time-consuming (39). CAR-NK cells demonstrate substantially lower rates of CRS and neurotoxicity and carry minimal risk of GvHD, enabling the development of allogeneic off-the-shelf products (30, 44). Unlike CAR-T cells, CAR-NK cells retain endogenous receptor-mediated cytotoxicity, allowing recognition and elimination of tumor cells even when target antigen expression is reduced or lost. However, CAR-NK cells generally exhibit shorter persistence and lower expansion capacity, which may limit durable tumor control (45).
Given their distinct biological characteristics, CAR-T and CAR-NK therapies may represent complementary rather than competing platforms; however, the clinical role of CAR-NK cells in GI malignancies remains investigational. CAR-T cells may be advantageous when prolonged persistence and memory formation are required, whereas CAR-NK cells may be preferable when safety, rapid availability, and resistance to antigen-loss escape are priorities.
2.5. Manufacturing and practical considerations
Practical aspects of CAR-based therapy manufacturing directly affect clinical outcomes and trial interpretation in GI oncology. Autologous CAR-T production requires individualized manufacturing with vein-to-vein times typically ranging from 4 to 6 weeks (46). In the pivotal CT041-ST-01 trial (NCT04581473), 16 of 104 patients randomized to satricabtagene autoleucel (15%) did not receive the product, primarily due to disease progression during the manufacturing period (29). This attrition rate is clinically significant in GI cancers, where rapid disease progression is common, and underscores the need for faster manufacturing processes or off-the-shelf alternatives.
Lymphodepleting conditioning regimens, most commonly fludarabine and cyclophosphamide, are administered before CAR-cell infusion to enhance engraftment and expansion. A substantial proportion of the grade 3 or higher hematologic toxicities reported across GI CAR-T trials including lymphopenia, neutropenia, and thrombocytopenia are attributable to lymphodepletion rather than the CAR-T product itself (29, 47). Distinguishing conditioning-related from product-related toxicity is essential for accurate safety assessment.
Response assessment in CAR-T trials also presents unique challenges. Delayed responses, pseudoprogression, and tumor enlargement between enrollment and infusion can reduce apparent ORR when assessed by conventional RECIST criteria. In the CT041-ST-01 trial, several patients demonstrated tumor growth during the manufacturing interval, which may have underestimated the true treatment effect (29). These considerations are important when comparing efficacy across different targets and trial designs. Cost remains a major barrier to broad implementation. Autologous CAR-T products typically exceed $300,000–$500,000 per patient, excluding hospitalization and supportive care costs. Geographic access is also limited, with the majority of GI CAR-T trials concentrated in China and the United States (47). Allogeneic CAR-NK platforms and emerging in vivo CAR-generation technologies are being investigated as potential approaches to address cost and access limitations, although their clinical role in GI malignancies remains to be established.
3. Target antigens for CAR-based therapy in GI malignancies
3.1. Overview of target selection in GI oncology
The therapeutic efficacy of CAR-based cellular therapies is fundamentally determined by the selection of an appropriate tumor-associated antigen. An ideal target should exhibit high, homogeneous surface expression on malignant cells while demonstrating minimal expression in normal tissues, thereby limiting on-target, off-tumor toxicity (48). Additional desirable characteristics include functional importance to tumor biology (making antigen loss biologically costly), stable membrane localization resistant to proteolytic shedding, accessibility to extracellular binding domains, and the availability of validated diagnostic assays for patient selection.
In GI malignancies, target selection is particularly challenging because many candidate antigens are co-expressed at low levels in normal epithelial tissues throughout the digestive tract (28). Intratumoral heterogeneity, temporal fluctuations in antigen expression, and antigen loss under therapeutic pressure further compound these difficulties, contributing to treatment resistance and disease recurrence (49, 50). Antigen escape, mediated through downregulation, epitope masking, or lineage switching, has been documented across multiple solid tumor CAR-T programs and remains a central obstacle to durable responses (51). Despite these challenges, six target antigens have now generated sufficient clinical evidence to permit meaningful evaluation of their therapeutic potential: CLDN18.2, GUCY2C, MSLN, GPC3, CEA, and HER2. Among these, CLDN18.2 is the most clinically advanced, with the first positive randomized controlled trial result for a CAR-based therapy in any solid tumor (29). Table 1 summarizes the clinical development status of these six targets.
Table 1.
Clinically advanced targets antigens for CAR-based cell therapy in GI oncology.
| Target | Main GI cancers | Platform | Highest phase | Key findings | Main limitations | References |
|---|---|---|---|---|---|---|
| CLDN18.2 | Gastric/GEJ, pancreatic | CAR-T | Phase II | CT041/satri-cel improved ORR and mPFS in advanced GC/GEJ | Antigen heterogeneity, limited persistence | (29, 52, 53) |
| GCC/GUCY2C | Metastatic CRC | CAR-T | Phase I | GCC19CART showed ORR 40% (China); DL2 ORR 80% in US cohort | Colitis/diarrhea, durability concerns | (54, 55) |
| MSLN | Pancreatic, gastric, biliary | CAR-T; CAR-NK (primarily preclinical/early-stage) | Phase I/II | Gavo-cel achieved ORR 20%, DCR 77% | Dense stroma, poor trafficking | (56–60) |
| CEA | CRC, gastric, pancreatic | CAR-T | Phase I | Early disease control; newer hypoxia-responsive CARs improved activity | On-target GI toxicity | (61–63) |
| GPC3 | HCC | CAR-T; CAR-NK (primarily preclinical/early-stage) | Phase I/II | IL-15-armored GPC3 CAR-T ORR 33%, DCR 66% | TGF-β-rich TME, limited persistence | (54, 64–66) |
| HER2 | Gastric/GEJ, biliary, CRC | CAR-T | Phase I/II | TAC01-HER2 DCR 61.1% (11/18); PRs observed in gastric/GEJ patients | Narrow therapeutic window | (67) |
CLDN18.2, Claudin 18.2; GCC/GUCY2C, Guanylyl Cyclase C; MSLN, mesothelin; CEA, carcinoembryonic antigen; GPC3, glypican-3; HER2, human epidermal growth factor receptor 2; CAR, chimeric antigen receptor; CAR-NK, chimeric antigen receptor natural killer; CRC, colorectal cancer; DCR, disease control rate; GC, gastric cancer; GEJ, gastroesophageal junction; GI, gastrointestinal; HCC, hepatocellular carcinoma; mPFS, median progression-free survival; ORR, objective response rate; PR, partial response; TGF-β, transforming growth factor beta.
Given the distinct biological characteristics of individual GI malignancies, we additionally summarize the relevant CAR targets and available evidence by tumor type to facilitate disease-specific interpretation Table 2.
Table 2.
Tumor type-specific overview of CAR-based targets and clinical evidence in GI malignancies.
| Tumor types | Relevant CAR target(s) | CAR platform | Clinical development/evidence | Key considerations |
|---|---|---|---|---|
| PDAC | MSLN, CEA, HER2, CLDN18.2 | Primarily CAR-T, CAR-NK/NKT approaches remain early-stage/preclinical | Early-phase clinical investigation, no established CAR-based therapy | Dense desmoplastic stroma, antigen heterogeneity, poor trafficking, and immunosuppressive TME limit efficacy |
| CRC | GUCY2C, CEA, HER2, MSLN | Primarily CAR-T, CAR-NK/NKT evidence remain limited | GUCY2C and CEA have entered early-phase clinical evaluation, regional delivery has also been investigated | Antigen heterogeneity, antigen escape, trafficking limitations, and potential on-target/off-tumor toxicity |
| Biliary tract cancer (BTC) | MSLN, CLDN18.2, HER2 | Primarily CAR-T, CAR-NK/NKT evidence remain limited | Predominantly preclinical and early-phase clinical investigation | Target expression varies by anatomical subtype; tumor heterogeneity and an immunosuppressive microenvironment remain important barriers |
| HCC | GPC3 | Primarily CAR-T, CAR-NK/NKT remains early-stage | Early-phase clinical studies of GPC3-directed CAR-T have shown antitumor activity; CAR-NK studies remain immature | GPC3 has relatively restricted expression in normal adult tissues; persistence, trafficking, and the immunosuppressive TME remain barriers |
| GEJ cancer | MSLN, CLDN18.2, HER2 | Primarily CAR-T | CLDN18.2 has the most advanced clinical evidence, including randomized phase II evidence with satri-cel | CLDN18.2 has a favorable tumor-associated expression profile; antigen heterogeneity and variable target expression remain relevant |
| Esophageal cancer | MSLN, HER2 | Primarily CAR-T | Limited clinical evidence; development remains predominantly early-stage | Variable target expression and tumor heterogeneity limit broad applicability |
BTC, biliary tract cancer; CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; CLDN18.2, claudin 18.2; CRC, colorectal cancer; GEJ, gastroesophageal junction; GPC3, glypican-3; HCC, hepatocellular carcinoma; HER2, human epidermal growth factor receptor 2; MSLN, mesothelin; NK, natural killer; NKT, natural killer T; PDAC, pancreatic ductal adenocarcinoma; TME, tumor microenvironment.
3.2. Target ANTIGENS
3.2.1. Claudin 18.2 (CLDN18.2)
3.2.1.1. Biological rationale and expression
CLDN18.2 is a tight-junction protein that controls epithelial barrier integrity and polarity (68, 69). Under normal conditions, its expression stays largely restricted to differentiated gastric mucosal cells, where tight junctions sequester it away from circulating immune cells (70). During malignant transformation, disruption of epithelial architecture exposes CLDN18.2 on the tumor cell surface, creating a therapeutic opportunity for antibody- and CAR-based targeting (70, 71).
CLDN18.2 appears most often in GEJ adenocarcinomas, with positivity rates from about 40% to 54% depending on the assay and scoring criteria (29, 72). Lower frequencies have also been reported in PDAC, esophageal adenocarcinoma, and selected CRC (73, 74). Because CLDN18.2 has a restricted normal-tissue distribution yet is broadly expressed across GI malignancies, it has become one of the most promising CAR targets. The success of the anti-CLDN18.2 antibody zolbetuximab in advanced gastric cancer further strengthened clinical validation (70).
3.2.1.2. Preclinical development
Early studies showed that CLDN18.2-targeted CAR-T cells have potent antitumor activity in gastric cancer xenograft and patient-derived xenograft models, producing durable tumor regression without detectable toxicity to normal gastric tissue (75). Later work focused on overcoming the solid-tumor microenvironment. IL-15-armored CLDN18.2 CAR-T cells improved expansion and persistence, and AZD6422 (which contains a dominant-negative TGF-β receptor) showed enhanced activity against CLDN18.2-positive gastric, pancreatic, and esophageal cancers (52, 76).
3.2.1.3. Clinical development of satricabtagene autoleucel
Satricabtagene autoleucel (satri-cel, CT041) has led the clinical development of CLDN18.2-directed CAR-T therapy. In the phase I CT041-CG4006 study (NCT03874897), Qi et al. (2022) reported an ORR of 57.1% and a disease control rate (DCR) of 75.0% among heavily pretreated gastric cancer patients (53). Updated results from 98 patients across GI malignancies showed an ORR of 38.8%, a DCR of 91.8%, a median PFS of 4.4 months, and a median OS of 8.8 months (77).
The international ELIMYN18.2 phase Ib trial (NCT04404595) supported these findings, reporting ORRs of 42.9% in gastric/GEJ cancer and 16.7% in pancreatic cancer, confirming activity beyond gastric malignancies (78).
The randomized phase II CT041-ST-01 trial (NCT04581473) provided the first randomized evidence of CAR-T efficacy in a solid tumor. Among 156 patients with previously treated CLDN18.2-positive gastric or GEJ cancer, satri-cel significantly improved median PFS compared with physician’s choice therapy (3.25 vs. 1.77 months; HR 0.37, 95% CI 0.24–0.56; p < 0.0001) (29). In the ITT population, OS favored satri-cel but did not reach statistical significance (7.92 vs. 5.49 months; HR 0.69, 95% CI 0.46–1.05), whereas the mITT population showed a significant OS difference (8.61 vs. 5.49 months; HR 0.60, 95% CI 0.38–0.94) (29). Crossover from physician’s choice therapy to satri-cel occurred in 20 patients (38%) and may have diluted the ITT OS estimate (29). These findings established CLDN18.2 as the first CAR-T target with randomized evidence of clinical benefit in a solid tumor.
3.2.1.4. Safety profile
Across clinical studies, satri-cel has shown a manageable but not negligible safety profile. In the CT041-ST-01 trial, CRS occurred in 95% (84/88) of treated patients but was predominantly grade 1–2; grade ≥3 CRS was uncommon. No cases of ICANS were reported. However, 99% of patients (87/88) experienced grade ≥3 treatment-emergent adverse events, including grade ≥3 lymphocyte decrease in 98%, white blood cell decreases in 77%, and neutrophil decrease in 66%, largely attributable to lymphodepleting chemotherapy (29). One treatment-related death (disseminated intravascular coagulation) occurred in the satri-cel arm (29). The phase I CT041-CG4006 trial (NCT03874897) reported no ICANS or treatment-related deaths (53, 77). Low-grade gastric mucosal injury was the main on-target, off-tumor toxicity and likely reflects transient CLDN18.2 exposure in normal gastric epithelium (77).
3.2.1.5. Comparison with other CLDN18.2-directed therapeutic modalities
Because both satri-cel and zolbetuximab target CLDN18.2, comparison between these modalities is clinically relevant, although cross-trial comparisons should be interpreted cautiously given their evaluation in different treatment settings. Zolbetuximab is approved in the first-line setting in combination with chemotherapy, whereas satri-cel has been studied primarily in previously treated disease. Key differences in therapeutic modality, eligibility criteria, efficacy outcomes, toxicity profiles, and logistical considerations are summarized in Table 3.
Table 3.
Comparison of CLDN18.2-targeted therapies in advanced gastric and gastroesophageal junction cancer.
| Feature | Satri-cel (CT041) | Zolbetuximab (Vyloy) | References |
|---|---|---|---|
| Therapeutic Modality | CLDN18.2-directed autologous CAR-T cell therapy | Anti-CLDN18.2 monoclonal antibody (IgG1) | (77, 79) |
| Regulatory Status | Investigational (phase II) | FDA-approved (June 2025); approved in Japan (March 2024) | (80) |
| Clinical Setting | Previously treated advanced gastric/GEJ cancer (≥2 prior systemic therapies) | First-line HER2-negative, CLDN18.2-positive advanced gastric/GEJ adenocarcinoma in combination with chemotherapy | (29, 81, 82) |
| CLDN18.2 Eligibility | ≥2+ staining in ≥40% of tumor cells | Moderate-to-strong staining in ≥75% of tumor cells | (29, 83) |
| Pivotal Trial(s) | CT041-ST-01 (NCT04581473) (phase II, randomized; n = 156) | SPOTLIGHT (NCT03504397; phase III; n = 565); GLOW (NCT03653507; phase III; n = 507) | (29, 84, 85) |
| Median PFS | 3.25 vs. 1.77 months; HR 0.37 (95% CI 0.24–0.56) | SPOTLIGHT: 10.6 vs. 8.7 months; HR 0.75. GLOW: 8.2 vs. 6.8 months; HR 0.69 | (29, 84, 85) |
| Median OS | 8.61 vs. 5.49 months (mITT); HR 0.60a | Combined SPOTLIGHT/GLOW analysis: 16.4 vs. 13.7 months; HR 0.77 | (29, 85) |
| Administration | Intravenous infusion following lymphodepleting chemotherapy; up to three infusions permitted | Intravenous infusion every 2–3 weeks until disease progression or unacceptable toxicity | (29, 80) |
| Manufacturing | Autologous product requiring leukapheresis and ex vivo cell engineering | Off-the-shelf biologic agent | (77, 80) |
| Manufacturing Time | Not reported for satri-cel; autologous CAR-T production generally requires several weeksb | Immediate availability | (29, 46) |
| CRS/Neurotoxicity | CRS in 95% of patients (predominantly grade 1–2); no ICANS reported | Not applicable | (29, 77) |
| Key Toxicities | Cytopenias largely related to lymphodepletion; low-grade CRS; occasional gastric mucosal injury | Nausea, vomiting, decreased appetite, and fatigue | (29, 84, 85) |
| Treatment-Related Deaths | One treatment-related death (disseminated intravascular coagulation) | Rarec | (29, 84, 85) |
| Treatment Duration | Time-limited cellular therapy | Continuous treatment until progression | (84, 85) |
| Major Advantages | Demonstrated efficacy in heavily pretreated disease; potential for durable cellular immune surveillance; broader eligibility criteria in clinical trials | Phase III evidence; immediate availability; no manufacturing delay | (29, 84, 85) |
| Major Limitations | Manufacturing delay; lymphodepletion required; CRS risk; currently supported by phase II data | Continuous administration; chemotherapy backbone required; narrower biomarker-defined population | (46, 77, 83) |
CAR, chimeric antigen receptor; CRS, cytokine release syndrome; GEJ, gastroesophageal junction; HR, hazard ratio; ICANS, immune effector cell-associated neurotoxicity syndrome; mITT, modified intention-to-treat; OS, overall survival; PFS, progression-free survival.
Median OS was reported in the modified intention-to-treat population (8.61 vs. 5.49 months; HR 0.60, 95% CI 0.38–0.94). CT041-ST-01 was powered for progression-free survival rather than overall survival, and crossover may have influenced OS analyses.
Satri-cel-specific manufacturing turnaround time has not been formally reported. The estimate reflects typical autologous CAR-T manufacturing timelines. Bridging therapy was permitted in CT041-ST-01.
Fatal adverse events occurred in phase III zolbetuximab studies, although investigator-attributed treatment-related deaths were uncommon.
This comparison is limited by fundamental differences in treatment setting (previously treated vs. first-line), trial design (phase II vs. phase III), patient populations, and CLDN18.2 eligibility thresholds. Direct comparison of efficacy outcomes is not appropriate.
This comparison highlights the different clinical roles currently occupied by CLDN18.2-directed therapies. Zolbetuximab is an off-the-shelf monoclonal antibody approved for first-line use with chemotherapy, while satri-cel is an investigational cellular therapy being developed for previously treated disease. Whether these approaches should be used sequentially, competitively, or in complementary settings remains an important question for future trials.
3.2.1.6. Current limitations and future directions
Despite its clinical progress, CLDN18.2-directed CAR-T therapy still faces several barriers. Antigen heterogeneity and antigen downregulation may contribute to resistance and relapse (76), while limited persistence, an immunosuppressive TME, and autologous manufacturing time may restrict long-term benefit (29, 52, 77). In CT041-ST-01 (NCT04581473), 16 of 104 patients randomized to satri-cel did not receive treatment, mainly because of disease progression during manufacturing, emphasizing the practical limitations of individualized CAR-T production (29).
Current strategies aim to improve persistence and resistance to tumor-mediated suppression through cytokine armoring, dominant-negative TGF-β receptors, dual-target constructs, affinity tuning, earlier-line treatment, and allogeneic platforms (52, 76). The growing CLDN18.2 therapeutic landscape also raises key sequencing questions, particularly whether prior zolbetuximab exposure affects CLDN18.2 expression or subsequent CAR-T efficacy. Preliminary data from a small subgroup of five patients receiving satri-cel sequentially after first-line therapy showed encouraging outcomes, but these findings require confirmation in larger studies (86). Overall, CLDN18.2 remains the most clinically advanced CAR target in GI oncology and a leading model for solid-tumor CAR development.
3.2.2. Guanylyl cyclase C (GCC/GUCY2C)
3.2.2.1. Biological rationale and expression
GUCY2C, also known as GCC is a transmembrane receptor normally expressed on the apical surface of intestinal epithelial cells, where it regulates epithelial homeostasis through cyclic GMP signaling (87, 88). During colorectal tumorigenesis, expression of its endogenous ligand guanylin is lost, leading to functional silencing of the pathway while receptor expression is largely retained (82, 89).
GUCY2C is expressed in approximately 95% of primary CRC and 70–80% of metastatic lesions, with expression generally maintained across disease stages and metastatic sites (55, 87). Importantly, its normal expression is largely confined to the luminal surface of intestinal epithelium, limiting exposure to circulating immune cells, whereas loss of epithelial polarity during malignant transformation exposes the receptor on tumor cells and creates a therapeutic window for CAR-based targeting (81, 90).
3.2.2.2. Preclinical development
Magee et al. (2016) first demonstrated that GUCY2C-directed CAR-T cells eradicated CRC metastases in syngeneic mouse models without inducing detectable intestinal autoimmunity (81). Subsequent studies confirmed potent antitumor activity in xenograft models and prolonged survival (90). Optimization efforts focused on CAR architecture, with CD8α hinge and transmembrane domains improving antigen sensitivity, cytokine production, persistence, and overall antitumor efficacy compared with CD28-based structural domains (91).
3.2.2.3. GCC19CART and the CoupledCAR platform
A major advance in the field has been the GCC19CART platform, which uses a CoupledCAR strategy in which GCC-directed CAR-T cells are co-administered with CD19-directed CAR-T cells that provide cytokine support following activation by normal B cells (Chen et al., 2024). This approach was designed to address one of the principal limitations of solid-tumor CAR-T therapy: inadequate in vivo expansion and persistence. However, the contribution of CD19-directed B-cell depletion and associated immune modulation to the observed antitumor responses, beyond simple cytokine support, remains incompletely understood and warrants further mechanistic investigation.
In the first-in-human phase I study, GCC19CART achieved an ORR of 40% among 15 evaluable patients with relapsed or refractory metastatic CRC, with all responses classified as partial responses (55). Clinical activity was dose dependent, with ORRs of 25% and 57% at dose levels 1 and 2, respectively. Median PFS improved from 1.9 to 6.0 months, while median OS increased from 13.3 to 22.8 months across dose levels (55, 92).
Preliminary results from a parallel US phase I study (NCT05319314) were similarly encouraging. Among nine treated patients, ORR reached 80% (4/5 patients) at dose level 2, including one pathological complete response and three confirmed partial responses, though this was based on a very small sample size (93). Although promising, these findings require confirmation in larger studies.
3.2.2.4. Safety profile
The most common toxicities have been CRS and GI adverse events. In both Chinese and US studies, CRS occurred in nearly all patients but was predominantly grade 1-2 (92, 93). Diarrhea was frequent and represented the major dose-limiting toxicity.
A notable safety signal emerged in the US study, where one patient developed grade 4 enterocolitis complicated by fungal sepsis resulting in treatment-related death (93). This event highlights the challenges of targeting an antigen that remains physiologically expressed within normal intestinal tissue despite its favorable compartmentalization.
3.2.2.5. Current limitations and future directions
Despite encouraging early activity, clinical evidence remains limited to small phase I studies. GI toxicity, particularly diarrhea and colitis, remains the major barrier to broader implementation. Additional concerns include heterogeneous antigen expression, limited long-term persistence, and the potential for adaptive antigen loss. Preclinical studies suggest that inflammatory signaling triggered during CAR-T activation may contribute to GUCY2C downregulation and therapeutic resistance (94).
Future efforts are likely to focus on improving toxicity management, enhancing persistence, and reducing antigen escape through next-generation CAR designs and multi-antigen targeting approaches. Given its highly restricted expression pattern and encouraging early clinical activity, GUCY2C remains one of the most promising targets for CAR-based therapy in metastatic CRC.
3.2.2.6. Additional GUCY2C-directed CAR-T programs
Beyond the GCC19CART CoupledCAR platform, a separate GUCY2C-directed CAR-T product has been evaluated in a phase I study (NCT05287165) (Qi et al., 2026). Among 19 evaluable patients with metastatic CRC, the overall ORR was 26.3% (5/19), with ORR increasing to 40% at dose level 3. Median PFS was 7.0 months at dose level 3. Notably, grade 3 diarrhea occurred in 55% of patients, confirming GI toxicity as the principal dose-limiting challenge for GUCY2C-targeted therapies regardless of CAR construct design (95). These results from a distinct GUCY2C program reinforce both the therapeutic potential and the safety challenges associated with this target.
3.2.3. Mesothelin
3.2.3.1. Biological rationale and expression profile
MSLN is a 40-kDa glycosylphosphatidylinositol (GPI)-anchored cell surface glycoprotein. It is derived from a 69–71 kDa precursor that splits into membrane-bound mesothelin and a soluble fragment called megakaryocyte-potentiating factor (MPF) (96). Normal expression is largely confined to mesothelial cells lining the pleura, peritoneum, and pericardium. Mesothelin drives tumor progression by interacting with MUC16/CA125 and activating the MSLN/Akt/NF-κB/IL-6/Mcl-1 signaling axis, which enhances tumor survival, invasion, and resistance to apoptosis (97, 98). Because of its high expression on tumors and limited distribution in normal tissues, mesothelin has become one of the most heavily investigated targets for adoptive cellular therapy in solid tumors.
3.2.3.2. Expression across gastrointestinal malignancies
PDAC shows the highest and most consistent mesothelin expression among GI cancers. Positivity has been reported in 80–100% of tumors, with minimal expression in normal pancreas or chronic pancreatitis (99). Weidemann et al. (2021) confirmed positivity in 89% of 599 PDAC, and BASECAMP-1 transcriptomic analyses found significantly higher MSLN expression in pancreatic cancer than in colorectal or lung cancer (100, 101). High mesothelin expression has also been linked to reduced cytotoxic T-cell infiltration (102).
In BTCs, expression varies by subtype. Extrahepatic bile duct adenocarcinomas show near-universal expression, whereas intrahepatic cholangiocarcinoma has lower positivity rates from 33% to 79% (99, 103, 104). Mesothelin expression is associated with metastatic behavior and worse outcomes in BTCs (105, 106).
Gastric cancer expresses mesothelin in roughly 49-59% of cases. Several studies have linked high expression to peritoneal recurrence and poor prognosis, though conflicting results exist (107–109). In gastroesophageal adenocarcinoma, expression rises with advancing stage (110).
CRC generally has lower and more heterogeneous expression. Most CRCs express little or no mesothelin, but about 10% show high-level expression comparable to ovarian cancer, especially among KRAS/BRAF-mutant tumors (111).
3.2.3.3. Preclinical development
Lv et al. (2019) developed a third-generation mesothelin-directed CAR (M28z10) containing CD3ζ, CD28, and DAP10 signaling domains. These CAR-T cells showed potent activity against mesothelin-positive gastric cancer and caused significant tumor regression in multiple xenograft models. Regional delivery worked better than systemic administration (112).
Targeting membrane-proximal mesothelin epitopes proved more effective. Tomar et al. (2022) showed that hYP218-based CAR-T cells outperformed earlier SS1-based constructs (113). Liu et al. (2024) then identified soluble mesothelin-mediated sequestration as a resistance mechanism and found that targeting a juxtamembrane epitope overcame this problem, producing complete regressions in pancreatic patient-derived xenograft models. Similar hYP218-based CAR-T cells eradicated cholangiocarcinoma xenografts with durable persistence (60, 103).
Researchers have designed multiple engineering strategies to tackle the immunosuppressive tumor microenvironment. Armstrong et al. (2025) created a CAR-TEAM platform where mesothelin-directed CAR-T cells secrete a fibroblast activation protein (FAP)-targeted T-cell engager, which improved activity when combined with ibrutinib and PD-1 blockade (114). Other approaches include proton radiation to raise mesothelin expression (115), oncolytic herpes simplex virus to induce mesothelin on antigen-negative tumor cells (116), and membrane-bound TRAIL-armored CAR-T cells that can kill both MSLN-positive and MSLN-negative tumor populations (117).
Preclinical CAR-NK approaches have also shown promise. Mesothelin-specific CAR NK-92 cells worked against gastric cancer and CRC models (111). IL-15-enhanced mesothelin-targeted CAR-NKT cells generated from hematopoietic stem cells achieved potent antitumor activity in orthotopic and metastatic PDAC models without GvHD (118).
3.2.3.4. Clinical development
Beatty et al. (2018) conducted a phase I study (NCT01897415) of mesothelin-directed CAR-T therapy. They gave mRNA-electroporated CARTmeso cells to six patients with chemotherapy-refractory metastatic PDAC. No CRS or dose-limiting toxicities occurred. Two patients had stable diseases, and one showed a 69.2% reduction in metabolically active tumor volume with complete resolution of FDG uptake in liver metastases (57).
Haas et al. (2019) reported the first lentiviral mesothelin CAR-T trial (NCT02159716) in 15 patients with advanced solid tumors, including five with PDAC. Eleven patients achieved stable disease, and CAR-T cells were found in tumor biopsies. But human anti-chimeric antibodies (HACAs) developed in more than half of treated patients, identifying murine SS1 scFv immunogenicity as a major limitation (58). Ko et al. (2020) tested dual CART-Meso/CART-19 therapy in metastatic PDAC but observed limited persistence and only modest clinical benefit (119).
The most mature efficacy data come from the phase 1/2 gavo-cel trial (NCT03907852), which used a T-cell receptor fusion construct (TRuC) targeting mesothelin (59). Unlike conventional CARs that signal through synthetic intracellular domains, TRuCs fuse the antigen-binding domain directly to the endogenous TCR complex, engaging all TCR signaling subunits. While mechanistically distinct from CAR technology, TRuC-based therapies are included here because they share the same therapeutic principle of redirecting T-cell specificity toward a defined tumor-associated antigen. Among 30 evaluable patients, the unconfirmed overall response rate was 20% (confirmed ORR 13%), the DCR was 77%, and the 6-month OS rate was 70%. Responses correlated with reductions in soluble mesothelin-related peptides, while non-responders showed increased PD-L1 and CD155 expression. These results, though modest, provide proof of concept for TRuC technology in GI cancers.
Barber-Rotenberg et al. (2026) reported phase I results (NCT03054298) of fully human huCART-meso in 20 patients with mesothelin-expressing malignancies. Stable disease was the best overall response, observed in 60% of patients (12/20). One patient achieved a maximum reduction in target tumor volume of 41%, and one patient demonstrated CAR-T persistence exceeding two years (56). This supports the feasibility of fully human constructs that avoid HACA-mediated limitations.
Early conference data from the armored product UCMYM802 showed an ORR of 28.6% and a DCR of 57.1% among seven evaluable patients with MSLN-positive tumors, including PDAC and cholangiocarcinoma. One patient with extrahepatic cholangiocarcinoma had a partial response lasting six months (120).
Emerging platforms include the logic-gated Tmod CAR-T A2B694 (121) and intraperitoneal mesothelin-directed CAR-T cells with a PD-1 decoy receptor for gastroesophageal peritoneal carcinomatosis (122).
3.2.3.5. Safety profile
Grade 3 or higher CRS occurred in approximately 25% of patients treated with gavo-cel (15% at the RP2D), with comparable rates reported for huCART-meso and UCMYM802 (56, 59, 120).
Pulmonary toxicity is a particular concern because of normal mesothelin expression on mesothelial surfaces. In the gavo-cel trial, grade 3 or higher pneumonitis happened in 16% of patients, and one dose-limiting grade 5 bronchoalveolar hemorrhage was reported (59). These events have pushed development of safety-engineered platforms like logic-gated Tmod CAR-T systems. By contrast, mRNA-based CARTmeso had an excellent safety profile but limited durability because of transient CAR expression (57).
3.2.3.6. Current limitations and future directions
Several barriers continue to limit mesothelin-directed therapies. Soluble mesothelin, produced by ADAM17/TACE-mediated shedding, can sequester CAR-T cells and reduce antitumor activity (60). Tumor heterogeneity, especially in CRC and gastric cancer, allows antigen-negative escape. Dense stromal barriers and immunosuppressive signaling also impair CAR-T infiltration and persistence (114). Early trials revealed immunogenicity of murine scFvs, which prompted development of fully human constructs (56, 58).
Future development is focusing on membrane-proximal epitope targeting, armored CARs, cytokine-secreting and checkpoint-resistant constructs, dual-targeting approaches, allogeneic CAR-NKT platforms, logic-gated safety systems, and regional delivery strategies. Biomarker-guided patient selection and rational combination therapies will also be needed. Despite decades of work, no mesothelin-directed CAR product has reached phase III trials, underscoring that meaningful clinical translation remains an ongoing challenge.
3.2.4. Glypican-3
3.2.4.1. Biological rationale and expression profile
GPC3 is a 66-kDa cosyGPI-anchored heparan sulfate proteoglycan that is highly expressed during fetal development but largely absent from normal adult tissues (123, 124). Functionally, GPC3 acts as a co-receptor for Wnt ligands and promotes canonical Wnt/β-catenin signaling, a pathway central to HCC development and progression (125). Immunohistochemical studies demonstrate GPC3 expression in approximately 70–80% of HCCs, whereas expression is generally absent in normal hepatocytes, cirrhotic liver, and benign hepatic lesions (123, 126).
Its high prevalence, membrane localization, and relative tumor specificity have established GPC3 as the most extensively investigated CAR target in HCC. Because GPC3 contributes to tumor growth and survival, antigen loss may also impose a biological fitness cost on tumor cells (124).
3.2.4.2. Preclinical development
Early studies demonstrated that GPC3-targeted CAR-T cells selectively eliminated GPC3-positive HCC cells while sparing antigen-negative cells (127). Constructs incorporating 4-1BB costimulation achieved significant tumor regression and prolonged survival in xenograft models. To enhance persistence, Batra et al. (2020) engineered CAR-T cells co-expressing IL-15 and IL-21, resulting in improved expansion, increased memory phenotypes, and superior antitumor activity (128).
Several combination approaches have also shown promise. Sorafenib enhanced GPC3 CAR-T efficacy in preclinical models through modulation of the TME and augmentation of antitumor immunity (129). Additional strategies include armored CARs and checkpoint-resistant constructs incorporating dominant-negative TGF-β receptors to counteract the immunosuppressive liver microenvironment.
3.2.4.3. Clinical development
The first-in-human phase I study by Shi et al. (2020) established the feasibility of GPC3-directed CAR-T therapy in advanced HCC. Among 13 patients with GPC3-positive disease, two achieved confirmed partial responses and one maintained prolonged stable disease exceeding 44 months. OS rates at 6 months, 1 year, and 3 years were 50.3%, 42.0%, and 10.5%, respectively. CRS occurred in most patients but was predominantly grade 1–2, and no grade 3–4 neurotoxicity was reported (64).
Subsequent next-generation platforms have demonstrated encouraging activity. RUNX3-engineered GPC3 CAR-T cells (CT017; NCT03980288) achieved an ORR of 16.7% and a DCR of 50% in heavily pretreated HCC patients (130). Preliminary data from the CT0180 (NCT04756648) program similarly reported partial responses and disease stabilization.
A major advance was the development of IL-15-armored GPC3 CAR-T cells (15.CAR). In the study by Steffin et al. (2025), armored constructs achieved an ORR of 33% and a DCR of 66%, outperforming conventional GPC3 CAR-T products (54). Improved expansion, persistence, and intratumoral activity provided clinical proof-of-concept that cytokine armoring can overcome major barriers to CAR efficacy in solid tumors.
Additional encouraging results have emerged from the BEACON study evaluating Ori-C101, an armored GPC3-directed CAR-T product delivered through hepatic arterial infusion (131). Preliminary data presented at ASCO 2026 data (not yet peer-reviewed) demonstrated an ORR of 50%, a DCR of 77.8%, and a median overall survival of 14.4 months. At the recommended phase II dose, ORR increased to 66.7%, including a durable complete response lasting more than 20 months.
3.2.4.4. GPC3 CAR-NK approaches
GPC3 is also being investigated as a target for CAR-NK therapy in HCC, although current evidence remains predominantly preclinical. Yu et al. (2018) developed GPC3-specific CAR-NK92 cells that demonstrated potent antitumor activity both in vitro and in vivo. Importantly, activity was maintained under hypoxic conditions and in the presence of TGF-β, suggesting potential advantages over CAR-T cells within the hostile liver TME (132).
More recent efforts have focused on improving persistence and tumor homing through stem-cell-derived NK platforms and chemokine receptor engineering. Multiple phase I studies evaluating GPC3-targeted CAR-NK therapies are currently ongoing, although mature efficacy data remain unavailable.
3.2.4.5. Safety profile
CRS remains the most common toxicity associated with GPC3-targeted cellular therapy. Most cases are low grade and manageable with standard supportive measures, although severe CRS and rare treatment-related deaths have been reported (64). In contrast, ICANS has been uncommon in reported CAR-T studies; however, clinical safety data for CAR-NK therapy in HCC remain limited. Hematologic toxicities are frequently observed but are largely attributable to lymphodepleting chemotherapy rather than the cellular products themselves.
3.2.4.6. Current limitations and future directions
Despite encouraging progress, important challenges remain. Approximately 20-30% of HCCs lack GPC3 expression, limiting eligibility for treatment (123). Soluble GPC3 may also act as a decoy antigen that reduces CAR-mediated tumor recognition (133). Furthermore, the immunosuppressive liver microenvironment promotes T-cell exhaustion through TGF-β signaling, regulatory T cells, myeloid-derived suppressor cells, and checkpoint pathways.
Future development is focused on armored CAR constructs, dual-target strategies, regional delivery approaches such as hepatic arterial infusion, and off-the-shelf CAR-NK platforms. Integration with tyrosine kinase inhibitors, immune checkpoint blockades, and locoregional therapies may further improve outcomes. Taken together, current evidence supports GPC3 as the most clinically advanced CAR target in HCC and one of the strongest targets across GI oncology.
3.2.5. Carcinoembryonic antigen
3.2.5.1. Biological rationale and expression profile
CEA; CEACAM5 is a heavily glycosylated glycosylphosphatidylinositol-anchored cell-surface protein of the immunoglobulin superfamily (134, 135). In normal tissues, CEA is primarily expressed on the apical surface of GI epithelial cells, limiting accessibility to circulating immune cells. During malignant transformation, loss of epithelial polarity redistributes CEA across the tumor cell surface, increasing its accessibility to antibodies and engineered cellular therapies (135, 136).
CEA is highly expressed in CRC and is also frequently detected in gastric, gastroesophageal, pancreatic, and esophageal cancers (137–139). However, residual expression in normal GI and pulmonary epithelia creates an important risk of on-target/off-tumor toxicity.
3.2.5.2. Preclinical development
Early studies demonstrated that CEA-directed CAR-T cells mediate antigen-specific cytotoxicity, with later-generation constructs incorporating costimulatory domains to improve expansion and effector function (140). In CEA-transgenic models, antitumor activity could be achieved, although inflammatory toxicities were observed and partially mitigated by IL-6 receptor blockade.
Several combination strategies have enhanced efficacy. Cha et al. (2021) demonstrated that CEA CAR-T cells combined with an anti-CEA–IL-2 immunocytokine and cyclophosphamide produced durable antitumor responses (141), while Chi et al. (2019) showed that recombinant IL-12 augmented activity against colorectal, pancreatic, and gastric cancer models (142). More recently, Banik et al. (2026) identified the tumor glycocalyx as a physical barrier to CAR-T engagement and demonstrated that enzymatic remodeling restored antigen accessibility and improved cytotoxicity (143).
CEA-directed CAR-NK approaches are also under development. Preclinical studies using CAR-NK-92 cells have demonstrated CEA-dependent killing of colorectal cancer cells, and newer constructs incorporate strategies to improve tumor homing and resistance to the suppressive TME.
3.2.5.3. Clinical development
The first clinical study of CEA-directed CAR-T therapy was reported by Thistlethwaite et al. (2017) in patients with advanced CEACAM5-positive malignancies. No objective responses were observed, and intensified lymphodepletion resulted in transient respiratory toxicity attributed to low-level CEA expression in pulmonary epithelium. Limited CAR-T persistence was a major challenge (140).
Regional delivery subsequently emerged as a strategy to improve safety and efficacy. In the phase I Hepatic IO for Metastases (HITM) trial, Katz et al. (2015) administered anti-CEA CAR-T cells through hepatic artery infusion for CEA-positive liver metastases (144). Treatment was feasible and well tolerated, with evidence of tumor necrosis, fibrosis, and reductions in serum CEA. The subsequent HITM-SIR study combining CAR-T cells with selective internal radiation therapy demonstrated similar feasibility and acceptable safety (145).
Systemic administration has also shown activity. Zhang et al. (2017) treated 10 patients with metastatic colorectal cancer using second-generation CEA CAR-T cells. Seven patients achieved stable disease, including two lasting more than 30 weeks, with manageable toxicity (146).
A more recent advance is the hypoxia-responsive CEA CAR-T product PC13. In heavily pretreated patients with CEA-positive solid tumors, Gao et al. (2026) reported DCR of 82.4% following intraperitoneal administration and 68.0% following intravenous administration. ORR were 23.5% and 8%, respectively, with responses enriched in tumors expressing high levels of CEA (61).
Logic-gated approaches are also advancing clinically. The Tmod platform combines a CEA-targeted activating receptor with an inhibitory receptor recognizing HLA-A*02, enabling selective killing of tumor cells with HLA loss of heterozygosity while sparing normal tissues (63). Early EVEREST-1 data evaluating A2B530 demonstrated a favorable safety profile and preliminary antitumor activity.
3.2.5.4. Safety profile
On-target/off-tumor toxicity remains the principal concern for CEA-targeted therapies because of physiological expression in GI and pulmonary epithelia. Early systemic studies reported transient respiratory toxicity (140), while GI toxicities such as diarrhea and colitis remain potential risks.
Regional delivery has generally demonstrated a more favorable safety profile, with hepatic artery infusion studies reporting no severe CRS, neurotoxicity, or grade 3–4 CAR-T-related adverse events (144, 147). Hypoxia-responsive CAR-T therapy has primarily been associated with low-grade CRS, although grade 3 diarrhea has been reported (61).
3.2.5.5. Current limitations and future directions
Despite broad expression, several challenges limit the efficacy of CEA-targeted cellular therapy. Expression levels vary among tumors, antigen accessibility may be restricted by epithelial architecture and the glycocalyx, and CAR-T cells must overcome poor trafficking, limited persistence, and the immunosuppressive TME (139, 143).
Future strategies focus on improving specificity and efficacy through regional delivery, hypoxia-responsive CARs, logic-gated platforms such as Tmod systems, armored CARs, dual-target approaches, and CAR-NK therapies. Together, these advances may help overcome the historical limitations of CEA-directed cellular therapy and expand its role in GI oncology.
3.2.6. Human epidermal growth factor receptor 2 (HER2/ERBB2)
3.2.6.1. Biological rationale and expression profile
HER2, encoded by ERBB2 is a transmembrane receptor tyrosine kinase of the EGFR family that promotes tumor proliferation, survival, invasion, and therapeutic resistance through activation of the RAS-MAPK, PI3K-AKT-mTOR, and JAK-STAT pathways (148). HER2 is an attractive CAR target because tumors with ERBB2 amplification often express high surface antigen density and standardized diagnostic testing is widely available. However, HER2 is not tumor specific. Low-level expression is present in normal epithelial tissues, including the lung, GI tract, kidney, and heart, creating a significant risk of on-target/off-tumor toxicity (149).
3.2.6.2. Expression across gastrointestinal malignancies
HER2 overexpression or ERBB2 amplification occurs across several GI cancers. GEJ adenocarcinomas demonstrate the highest prevalence, with HER2 positivity in approximately 15-20% of cases, particularly in intestinal-type and proximal tumors (83, 150). In CRC, HER2 alterations occur in approximately 2-6% of tumors and are enriched in RAS wild-type disease, where they contribute to resistance to anti-EGFR therapy (151). HER2 alterations are also observed in BTC, especially gallbladder cancer and extrahepatic cholangiocarcinoma (152). By contrast, HER2-positive pancreatic cancer is relatively uncommon (153, 154). A major challenge for HER2-targeted therapy in gastric cancer is intratumoral heterogeneity, with patchy or incomplete membrane staining creating opportunities for antigen-negative escape (83).
3.2.6.3. Preclinical development
The development of HER2-directed CAR-T therapy has focused on balancing antitumor efficacy with safety. Early trastuzumab-derived CARs demonstrated potent cytotoxicity but raised concerns regarding recognition of low-density HER2 on normal tissues. Affinity-tuning approaches subsequently demonstrated that lower-affinity CARs could preferentially recognize HER2-high tumor cells while sparing HER2-low normal tissues (155, 156).
The Baylor group developed a lower-affinity FRP5-based HER2 CAR that became one of the most extensively studied clinical platforms (157, 158). More recent strategies include multiplex-engineered iPSC-derived HER2 CAR-T cells with enhanced persistence and resistance to TGF-β-mediated suppression (159). HER2-directed CAR-NK platforms have also shown promising preclinical activity and are being investigated as a potential alternative platform with distinct safety characteristics related to their shorter persistence (160, 161).
3.2.6.4. Clinical development
Clinical development of HER2-directed CAR-T therapy was profoundly influenced by the fatal toxicity reported by Morgan et al. (2010) (162). A patient with metastatic CRC rapidly developed respiratory failure and fatal cytokine release following infusion of a high-dose third-generation HER2 CAR-T product. This event was attributed to recognition of low-level HER2 expression in normal lung epithelium and remains one of the defining examples of on-target/off-tumor toxicity in solid tumor CAR therapy.
Subsequent programs adopted lower-affinity binding domains, reduced cell doses, and more conservative clinical designs. In GI cancers, Feng et al. (2018) evaluated HER2 CAR-T cells (NCT01935843) in 11 patients with HER2-positive biliary tract or pancreatic cancers. One partial response and five cases of stable disease were observed, with a median PFS of 4.8 months. Toxicities included grade 3 febrile syndrome and one severe GI hemorrhage at a tumor invasion site (163).
A more recent approach is TAC01-HER2 (NCT04727151), a T-cell antigen coupler platform that engages the endogenous T-cell receptor complex rather than conventional CAR signaling. In a phase I study including gastric and GEJ cancers, TAC01-HER2 achieved a DCR of 61.1%, with partial responses observed in gastric and GEJ adenocarcinoma patients and no treatment-related deaths (164).
Clinical experience with HER2 CAR-NK therapy remains limited. However, a phase I first-in-human study of HER2-targeted CAR-NK cells in recurrent glioblastoma demonstrated clinical feasibility and a favorable early safety profile, with no dose-limiting toxicities, cytokine release syndrome, or neurotoxicity reported favor (165).
3.2.6.5. Safety profile
Safety remains the major challenge for HER2-directed cellular therapy. The fatal toxicity reported by Morgan et al. (2010) highlighted the risks associated with high-affinity HER2 recognition, intensive lymphodepletion, and physiological HER2 expression in normal tissues (162). Later-generation approaches incorporating affinity tuning and modified receptor architecture have substantially improved safety.
Across more recent studies, CRS has generally been manageable, while ICANS has been uncommon. Cardiotoxicity remains a theoretical concern because HER2 is expressed in cardiac tissue, although clinically significant CAR-related cardiac toxicity has not been consistently observed. GI bleeding and inflammation may occur, particularly when tumors invade normal tissues (163).
3.2.6.6. Current limitations and future directions
Several barriers continue to limit HER2-directed cellular therapy in GI cancers. These include intratumoral HER2 heterogeneity, particularly in gastric and GEJ adenocarcinoma; the relatively low prevalence of HER2-positive disease in colorectal, pancreatic, and many biliary cancers; poor trafficking into solid tumors; and the immunosuppressive TME (83, 153).
Future strategies focus on affinity-tuned CARs, logic-gated receptor systems, multiplex-engineered iPSC-derived products, and HER2-directed CAR-NK platforms to improve both safety and efficacy (155, 159, 160). Dual-target approaches may also help address antigen heterogeneity. Although clinical efficacy remains modest, HER2 remains one of the most clinically validated oncogenic targets in GI oncology and a promising candidate for next-generation cellular therapies. Representative clinical trials of CAR-based cell therapies in GI malignancies are summarized in Table 4.
Table 4.
Representative clinical trials of CAR-based cell therapy in gastrointestinal malignancies.
| Target | Product | N | Cancer | Phase | Key findings | Main toxicity | References |
|---|---|---|---|---|---|---|---|
| CLDN18.2 | Satri-cel (CT041) | 88 (mITT) | Gastric/GEJ | Phase II | ORR 26% (IRC, mITT); mPFS 3.25 vs. 1.77 mo (HR 0.37) | CRS mainly G1–2 | (29) |
| CLDN18.2 | CT041 | 98 | GI cancers | Phase I | ORR 38.8%; GC ORR 57.1%; mPFS 4.2 months |
Mild CRS; no ICANS | (53) |
| GCC/GUCY2C | GCC19CART | 15 | Metastatic CRC | Phase I | ORR 40%; DL2 ORR 57%; mPFS 6.0 months (DL2) vs 1.9 months (DL1) | CRS; diarrhea/colitis | (55) |
| GCC/GUCY2C | GCC19CART | 9 | Metastatic CRC | Phase I | DL2 ORR 80% (4/5 pts) in US cohort; mPFS 7.8 months | CRS; ICANS; diarrhea | (93) |
| MSLN | Gavo-cel | 30 | Solid tumors incl. GI | Phase I/II | ORR 20%; DCR 77%; mPFS 5.6 months | CRS; pneumonitis | (59) |
| MSLN | CARTmeso | 6 | Pancreatic cancer | Phase I | ORR 0%; SD 33% (2/6); PFS 3.8 and 5.4 mo in patients with SD | Well tolerated | (57) |
| GPC3 | CAR-GPC3 | 13 | HCC | Phase I | ORR 15.4% (2/13 PR); 1 sustained SD; PFS 111 and 99 days in the two patients with PR | CRS | (64) |
| GPC3 | 15.CAR | 6 | GPC3+ tumors | Phase I | ORR 33%; DCR 66% | Increased CRS with expansion | (54) |
| HER2 | TAC01-HER2 | 18 | Gastric/GEJ-enriched | Phase I/II | ORR 11.1% (2/18); DCR 61.1% (11/18); mPFS 2.6 mo | CRS; pneumonitis | (166) |
| HER2 | CART-HER2 | 11 | BTC/pancreatic | Phase I | ORR 9.1% (1/11 PR); SD 45.5% (5/11); mPFS 4.8 mo | Febrile syndrome | (163) |
| CEA | Anti-CEA CAR-T | 10 | Metastatic CRC | Phase I | ORR 0%; SD 70% (7/10); median PFS NA | Minimal severe toxicity | (146) |
| CEA | PC13 | 42 | CEA+ solid tumors | Phase I | ORR 23.5% (IP) and 8.0% (IV); DCR 82.4% (IP) and 68.0% (IV); median PFS NA | Diarrhea; CRS | (61) |
4. Comparative analysis of clinically advanced CAR targets in gastrointestinal malignancies
Although several CAR-based therapies have entered clinical testing across GI malignancies, meaningful differences have emerged in their level of clinical maturity, efficacy signals, and safety profiles. Among currently investigated targets, CLDN18.2 has clearly advanced the furthest. The randomized phase II CT041-ST-01 trial established satricabtagene autoleucel (satri-cel) as the first CAR-T therapy to demonstrate a PFS benefit in a solid tumor, representing an important milestone for the field (29). In contrast, GUCY2C and GPC3 remain in earlier stages of development but have shown encouraging activity in metastatic CRC and HCC respectively (54, 55, 95, 131). The varying degrees of clinical success observed across targets appear to reflect fundamental differences in target biology. CLDN18.2 and GPC3 possess relatively restricted expression patterns in normal adult tissues, creating a more favorable therapeutic window and reducing the likelihood of severe on-target, off-tumor toxicity (29, 54). Similarly, GUCY2C remains an attractive target because its expression is largely confined to the apical surface of intestinal epithelial cells under normal physiological conditions while being retained in most CRC (167, 168). By comparison, mesothelin, CEA, and HER2 are expressed to varying degrees in normal tissues, increasing the potential for off-tumor toxicities and complicating clinical development (169, 170). Mesothelin presents an additional challenge because soluble shed mesothelin can act as a decoy antigen, limiting effective CAR-T cell engagement with tumor cells (60).
Another important theme emerging from clinical studies is the influence of delivery strategy. Conventional intravenous administration may be insufficient to overcome the anatomical and stromal barriers characteristic of many GI malignancies. Consequently, several groups have explored regional delivery approaches. Hepatic artery infusion of GPC3-directed CAR-T cells has generated encouraging response rates in advanced HCC (131), while intraperitoneal administration of CEA-targeted CAR-T cells has demonstrated greater activity than intravenous delivery in patients with peritoneal disease (61). These findings suggest that optimizing the route of administration may be as important as target selection in maximizing therapeutic efficacy.
Clinical experience also highlights the growing importance of next-generation engineering strategies. Early CAR constructs often demonstrated limited persistence and insufficient activity within the hostile tumor microenvironment. More recent approaches have incorporated cytokine armoring, resistance to immunosuppressive signaling pathways, and enhanced trafficking mechanisms. Notably, IL-15-armored GPC3 CAR-T cells achieved higher response rates than earlier generations of GPC3-directed products, providing one of the clearest clinical examples that rational engineering can improve outcomes in solid tumors (54). Similar strategies are now being explored for CLDN18.2, Mesothelin, and other GI cancer targets. Taken together, current evidence places CLDN18.2 at the forefront of clinical translation in GI oncology, supported by randomized evidence and multiple independent development programs. GUCY2C and GPC3 have emerged as particularly promising second-wave targets with encouraging early-phase efficacy signals, whereas Mesothelin, CEA, and HER2 continue to face important biological and safety challenges. As clinical development progresses, successful CAR-based therapy in GI cancers will likely depend on the combination of optimal target selection, innovative engineering approaches, and disease-specific delivery strategies rather than any single factor alone.
5. Challenges and barriers limiting CAR-based therapy in gastrointestinal malignancies
CAR-engineered immune cells have achieved remarkable success in hematologic cancers, but their efficacy in gastrointestinal solid tumors remains limited. GI cancers present a uniquely hostile environment: antigen heterogeneity, poor trafficking, profound immunosuppression, treatment-related toxicities, and manufacturing hurdles all play a role. These barriers together restrict the ability of CAR-T and CAR-NK cells to produce durable clinical responses (47, 170).
Figure 2. Barriers to CAR-T and CAR-NK cell efficacy in gastrointestinal solid tumors and emerging engineering solutions: The left panel illustrates the major biological barriers organized across three concentric zones: the vascular barrier (impaired transendothelial migration, chemokine–receptor mismatch, elevated interstitial fluid pressure), the stromal barrier (dense collagen and hyaluronic acid matrix, cancer-associated fibroblasts), and the immunosuppressive tumor core (antigen heterogeneity and escape, regulatory T cells, myeloid-derived suppressor cells, M2-polarized tumor-associated macrophages, soluble inhibitory mediators including TGF-β and adenosine, decoy antigens, T-cell exhaustion, and metabolic stress from hypoxia and nutrient depletion). The right panel summarizes the principal engineering strategies being developed to overcome each barrier, including chemokine receptor engineering, regional delivery, multi-antigen targeting, armored CAR constructs, persistence engineering, logic-gated safety systems, and scalable allogeneic manufacturing platforms.
Figure 2.

Barriers to CAR-T and CAR-NK cell efficacy in gastrointestinal solid tumors and emerging engineering solutions.
5.1. Antigen heterogeneity and antigen escape
A major challenge in GI malignancies is the lack of truly tumor-specific antigens. Most currently targeted molecules, including CLDN18.2, mesothelin, GPC3, HER2, CEA, and GUCY2C, are tumor-associated rather than tumor-specific and often display heterogeneous expression both between and within tumors (26, 47). This heterogeneity is particularly pronounced in PDAC and CRC, where antigen-negative or antigen-low subclones can survive CAR-mediated killing and drive disease recurrence (27). In addition, selective pressure from CAR therapy may induce antigen downregulation or complete antigen loss, resulting in acquired resistance similar to that observed with CD19-targeted therapies in hematologic malignancies (49).
Next-generation platforms are tackling these limits with tandem, bispecific, and pooled CAR strategies that recognize multiple antigens at once. Logic-gated systems (synNotch, split CARs) add precision by requiring multiple activation signals, and switchable CAR platforms allow retargeting of evolving antigen landscapes without making entirely new products (26, 171, 172).
5.2. Trafficking and infiltration barriers
Effective therapy requires CAR cells to reach and penetrate tumor tissue, yet this process is impaired at multiple levels in GI cancers. Abnormal tumor vasculature, characterized by disorganized endothelium, reduced expression of adhesion molecules such as VCAM-1 and ICAM-1, and elevated interstitial fluid pressure, limits transendothelial migration and intratumoral accumulation of CAR cells (172). In PDAC, the dense desmoplastic stroma composed of collagen, hyaluronic acid, and cancer-associated fibroblasts (CAFs) forms a substantial physical barrier to immune-cell infiltration (27, 173). In addition, chemokine-receptor mismatches frequently impair efficient homing of CAR cells to tumor sites.
Several approaches are being explored to improve trafficking, including engineering CAR cells to express tumor-matched chemokine receptors, vascular normalization with anti-VEGF therapies, stromal depletion through FAP-targeted approaches, and extracellular matrix degradation through heparanase expression (172, 173). Regional delivery has also emerged as a promising strategy. Hepatic artery infusion, intraperitoneal administration, and other locoregional approaches have demonstrated improved tumor exposure and enhanced antitumor activity while reducing systemic toxicity, particularly in liver and peritoneal metastases (61, 122, 170).
5.3. Immunosuppressive tumor microenvironment
TME of GI cancers is among the most immunosuppressive of all solid tumors. Regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSCs), tumor-associated macrophages (TAMs), and CAFs collectively suppress effector immune responses and frequently outnumber cytotoxic lymphocytes within tumors (27, 47). M2-polarized TAMs contribute further by increasing PD-L1 expression and reducing endothelial ICAM-1, thereby impairing CAR-cell trafficking and function (174).
Additional inhibitory mechanisms include checkpoint signaling through PD-1, PD-L1, CTLA-4, LAG-3, and TIGIT, as well as soluble suppressive mediators such as TGF-β, IL-10, adenosine, and prostaglandin E2 (27, 122). Metabolic stress caused by hypoxia, nutrient depletion, and acidic pH further compromises CAR-cell fitness and persistence (175).
To overcome these barriers, armored CAR platforms have been developed that secrete pro-inflammatory cytokines, express dominant-negative TGF-β receptors, incorporate checkpoint-switch receptors, or locally deliver checkpoint-blocking molecules. Additional strategies include CAR-macrophages and rational combinations designed to remodel the suppressive TME and restore effective antitumor immunity (176).
5.4. T-cell exhaustion and limited persistence
Persistent antigen exposure within solid tumors drives progressive CAR-T cell exhaustion, characterized by upregulation of inhibitory receptors such as PD-1, TIM-3, LAG-3, and TIGIT, accompanied by loss of proliferative capacity, diminished cytokine production, and reduced cytotoxicity (172). This dysfunctional state is reinforced by transcriptional regulators including TOX, NR4A family members, and REGNASE-1, which promote exhaustion-associated gene programs (172).
Current strategies to improve persistence include disruption of exhaustion-associated pathways through gene editing, enrichment of memory-like T-cell populations during manufacturing, optimization of co-stimulatory domains, and cytokine support through IL-15 or IL-21 engineering (177, 178). Similar approaches are being explored for CAR-NK cells, where limited in vivo persistence remains a major challenge.
5.5. On-target/off-tumor toxicity
Because most GI cancer antigens are also expressed at low levels in normal tissues, on-target/off-tumor toxicity remains a major safety concern. Clinical examples include fatal pulmonary toxicity following HER2-directed CAR-T therapy, colitis associated with CEA-targeted CAR-T cells, GI injury from CLDN18.2-targeted therapies (169, 170, 179).
Several engineering strategies have been developed to improve safety. These include affinity-tuned CARs that preferentially recognize high-density tumor antigens, logic-gated systems requiring multiple antigen inputs, inhibitory CARs, HLA-based discrimination approaches, suicide switches, and regional delivery methods that reduce systemic exposure (169–171). Beyond organ-specific toxicity, CRS remains the most common acute adverse event associated with CAR-T therapy, although it is generally less severe in GI cancer studies than in hematologic malignancies. CAR-NK have biological features that may confer safety advantages, including a lower propensity for CRS and a reduced risk of GvHD; however, clinical safety data in GI malignancies remain limited (176, 180).
5.6. Manufacturing challenges and scalability
Manufacturing remains a major obstacle to the widespread implementation of CAR-based therapies. Autologous products require individualized production, resulting in prolonged vein-to-vein times (typically 4–6 weeks), high costs (often exceeding $300,000–$500,000 per patient), and occasional manufacturing failure, particularly in heavily pretreated patients with compromised T-cell fitness (47). These challenges are especially relevant in GI cancers, where disease progression during manufacturing may preclude treatment.
Allogeneic platforms are being investigated as potential alternatives to autologous CAR-T manufacturing. CAR-NK cells derived from cord blood, NK-92 cell lines, or induced pluripotent stem cells (iPSCs) may enable standardized, large-scale manufacturing with a lower risk of GvHD (176, 177). Emerging in vivo CAR-generation technologies using lipid nanoparticles or viral vectors may further simplify production by eliminating the need for ex vivo manufacturing altogether (181). Nonviral gene-transfer systems such as Sleeping Beauty transposons and minicircle DNA vectors also provide lower-cost alternatives with reduced concerns regarding insertional mutagenesis.
Collectively, these challenges highlight why the success of CAR-based therapies in GI malignancies has lagged behind that observed in hematologic cancers. Future progress will likely depend on integrated solutions that simultaneously address antigen heterogeneity, trafficking, immunosuppression, toxicity, persistence, and manufacturing constraints. The major biological, safety, and practical barriers limiting CAR-based cell therapy in GI malignancies are summarized in Table 5.
Table 5.
Major barriers limiting CAR-T and CAR-NK therapy in GI solid tumors.
| Barrier | Mechanism/examples | Impact on CAR therapy | Principal GI cancers | Predominant platform | Emerging strategies | References |
|---|---|---|---|---|---|---|
| Antigen heterogeneity | Mosaic target expression across tumors/metastases; e.g., CLDN18.2, CEA, MSLN variability | Residual antigen-negative clones survive | Gastric, CRC, PDAC | Both | Tandem/multi-antigen CARs; logic-gated circuits; vaccine-boosted epitope spreading | (53, 182) |
| Antigen escape | Antigen downregulation/loss after immune pressure | Relapse after initial response | All GI cancers | CAR-T | Bispecific CARs; armored CARs; CAR-NK dual killing | (63, 183) |
| On-target/off-tumor toxicity | Target expression on normal tissues, especially GI epithelium/lung | Dose-limiting mucosal or organ toxicity | Gastric, CRC, HCC | CAR-T | Affinity-tuned CARs; Tmod/logic-gated systems; safety switches; hypoxia-responsive CARs | (61, 63, 140) |
| Dense desmoplastic stroma | CAFs, collagen, hyaluronan, high interstitial pressure | Poor tumor penetration | PDAC, CRC | Both | FAP-targeting strategies; heparanase-expressing CAR-T; regional delivery | (184, 185) |
| Poor trafficking/infiltration | Abnormal vasculature, chemokine mismatch, immune-excluded tumors | CAR cells fail to reach tumor core | PDAC, HCC, CRC | Both | CXCR2/CCR2b engineering; anti-VEGF combinations; HAI/IP delivery | (61, 184) |
| Immunosuppressive TME | Tregs, MDSCs, TAMs, TGF-β, IL-10, PGE2, PD-L1 | CAR-cell dysfunction after infiltration | All GI cancers | Both | IL-12/IL-18 armored CARs; TGF-β resistance; PD-1 KO/checkpoint blockade | (186, 187) |
| T-cell exhaustion | Chronic antigen stimulation; PD-1, LAG-3, TIM-3; NR4A/TOX-driven dysfunction | Reduced cytotoxicity and expansion | All GI cancers | CAR-T | NR4A/TOX targeting; PRDM1/NR4A3 modulation; optimized costimulation | (55, 188) |
| Metabolic suppression/hypoxia | Hypoxia, acidic pH, nutrient depletion, adenosine accumulation | Impaired survival and effector function | PDAC, HCC, CRC | Both | Hypoxia-responsive CARs; metabolic reprogramming; adenosine pathway targeting | (61, 183) |
| Soluble antigen/sEV shedding | Shed GPC3/HER2 or tumor-derived sEVs act as decoys | CAR sequestration and impaired activation | HCC, gastric | CAR-T | Membrane-proximal epitope targeting; sEV blockade; non-shed epitopes | (133, 187) |
| Limited persistence | Short in vivo survival and weak memory formation | Transient disease control | All GI cancers | CAR-NK | IL-15/IL-21 armoring; IL-7/CCL19 enrichment; repeat dosing | (64) |
| Allogeneic rejection | Host immune rejection via HLA mismatch and NK/T-cell recognition | Shorter persistence of off-the-shelf products | All GI cancers | Allogeneic CAR-NK | B2M KO + HLA-E; PD-L1 expression; CD54/CD58 ablation; iPSC platforms | (189, 190) |
CAFs, cancer-associated fibroblasts; CAR, chimeric antigen receptor; CRC, colorectal cancer; ECM, extracellular matrix; HCC, hepatocellular carcinoma; MDSCs, myeloid-derived suppressor cells; PDAC, pancreatic ductal adenocarcinoma; sEVs, small extracellular vesicles; TAMs, tumor-associated macrophages; TGF-β, transforming growth factor beta; TME, tumor microenvironment; Tregs, regulatory T cells.
6. Future directions
Despite substantial advances in target selection, durable responses to CAR-based therapies in GI malignancies remain uncommon. Current research is therefore shifting from simple antigen targeting toward more sophisticated engineering approaches designed to overcome antigen heterogeneity, immune suppression, poor trafficking, limited persistence, and manufacturing constraints. The principal strategies currently shaping the next generation of CAR therapies are summarized in Table 6.
Table 6.
Next-generation engineering strategies enhancing CAR therapy in GI oncology.
| Strategy | Intended benefit | Example GI application | Development stage | Key references |
|---|---|---|---|---|
| Armored/TRUCK CARs | Overcome immunosuppressive TME and enhance endogenous immune activation | CLDN18.2 CAR-T with dominant-negative TGFβRII (AZD6422) in gastric/pancreatic cancer | Preclinical; Phase I ongoing | (52) |
| IL-15/IL-21 engineered CARs | Improve persistence, memory formation, and resistance to exhaustion | IL-15-armored GPC3 CAR-T in HCC; MSLN CAR-iNK + IL-15 for solid tumors | Phase I (HCC); preclinical (solid tumors) | (54, 191) |
| Tandem/dual-target CARs | Reduce antigen escape and broaden tumor coverage | CLDN18.2 + NKG2DL CAR constructs in gastric/pancreatic cancer | Preclinical–early clinical | (53) |
| Logic-gated/Tmod CARs | Improve tumor specificity and reduce on-target/off-tumor toxicity | CEA/MSLN-targeting systems exploiting HLA LOH in GI tumors | Early clinical | (192–195) |
| SynNotch CARs | Enable spatially restricted CAR activation and local ECM remodeling | synNotch-induced CAR activation systems for solid tumors | Preclinical | (196) |
| Checkpoint-resistant CARs | Resist PD-L1-mediated exhaustion and suppressive signaling | PD-1–modified MSLN CAR-T for gastroesophageal/peritoneal disease | Preclinical; Phase I ongoing | (122) |
| Hypoxia-responsive CARs | Restrict CAR activation to hypoxic tumor regions and reduce systemic toxicity | PC13 hypoxia-responsive CEA CAR-T in CEA+ GI tumors | Phase I | (61, 197) |
| Chemokine receptor engineering | Improve tumor trafficking and intratumoral infiltration | CCR2b/CXCR2-engineered CAR-T in pancreatic/CRC models | Preclinical | (198) |
| Heparanase-expressing CAR-T | Enhance stromal penetration through ECM degradation | Heparanase-engineered CAR-T for desmoplastic PDAC models | Preclinical | (184) |
| Memory-enhanced CARs | Promote durable persistence and long-term antitumor activity | Tscm/Tcm-enriched CAR-T manufacturing approaches | Preclinical–early clinical | (188, 199) |
| Allogeneic off-the-shelf CAR-NK | Enable rapid scalable therapy with reduced CRS/GvHD risk | NKG2D CAR-NK in metastatic CRC | Phase I | (51) |
| iPSC-derived CAR-NK | Provide renewable, standardized, multi-engineered NK products | MSLN CAR-IL15 iNK platforms for solid tumors | Preclinical (GI); Phase I (hematologic malignancies) | (191, 200) |
| Regional delivery (IP/HAI) | Increase local tumor exposure while limiting systemic toxicity | IP CEA CAR-T for peritoneal carcinomatosis; HAI CAR-T for liver metastases | Phase I | (122, 144, 147) |
CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; CLDN18.2, claudin 18.2; CRC, colorectal cancer; CRS, cytokine release syndrome; CXCR2, C-X-C chemokine receptor 2; ECM, extracellular matrix; GI, gastrointestinal; GPC3, glypican-3; GvHD, graft-versus-host disease; HAI, hepatic artery infusion; HCC, hepatocellular carcinoma; HLA LOH, human leukocyte antigen loss of heterozygosity; IL, interleukin; iNK, induced natural killer; IP, intraperitoneal; iPSC, induced pluripotent stem cell; MSLN, mesothelin; NKG2DL, natural killer group 2D ligand; PDAC, pancreatic ductal adenocarcinoma; PD-L1, programmed death-ligand 1; Tcm, central memory T cell; TGFβRII, transforming growth factor beta receptor II; TME, tumor microenvironment; TRUCK, T cell redirected for universal cytokine killing; Tscm, stem cell memory T cell.
6.1. Precision antigen targeting: multi-antigen and logic-gated architectures
Antigen heterogeneity and antigen escape remain major barriers to effective CAR therapy in GI cancers. Tandem and dual-target CARs allow simultaneous recognition of multiple tumor-associated antigens, reducing the likelihood of relapse through loss of a single target antigen (171, 201). Preclinical studies in GI malignancies have demonstrated enhanced activity with constructs co-targeting combinations such as CLDN18.2/CD47 and GPC3/CD133.
Logic-gated systems provide an additional level of precision by integrating multiple antigen-recognition signals before CAR activation occurs. SynNotch platforms, for example, use an inducible circuit in which recognition of one antigen triggers expression of a second CAR, thereby improving tumor specificity. This approach has shown encouraging preclinical activity in HCC using GPC3-synNotch-inducible CD147 CAR-NK cells (32). Similarly, Tmod platforms combine activating and inhibitory receptors to distinguish malignant cells from normal tissues based on HLA loss-of-heterozygosity patterns. Early clinical evaluation of Tmod-based CAR-T cells targeting mesothelin in pancreatic and colorectal cancers is currently underway (NCT06051695) (121). Collectively, these approaches aim to improve efficacy while reducing on-target/off-tumor toxicity.
6.2. Armored CARs and persistence engineering
The highly immunosuppressive TME of GI cancers limits the activity and persistence of transferred immune cells. To address this challenge, fourth generation “armored” CARs, also known as TRUCKs, have been engineered to actively modify the local immune environment. These constructs secrete cytokines such as IL-12, IL-15, IL-18, or IL-21 following antigen engagement, promoting recruitment and activation of endogenous immune cells and enhancing antitumor immunity (202).
Among these strategies, IL-15 armoring has emerged as particularly important for CAR-NK therapies because it significantly improves in vivo persistence and expansion (30, 203). Additional approaches focus on overcoming inhibitory signaling pathways within the TME. Dominant-negative TGF-β receptor constructs render CAR cells resistant to TGF-β-mediated suppression, while PD-1-based switch receptors convert inhibitory signals into activating signals (202). Clinical translation of these approaches is already underway, including intraperitoneal mesothelin-targeted CAR-T cells incorporating PD-1 dominant-negative receptor technology for gastroesophageal cancer with peritoneal carcinomatosis (122).
Improving persistence also requires addressing T-cell exhaustion. Emerging strategies include CRISPR-mediated disruption of exhaustion-associated regulators such as TOX, NR4A, and REGNASE-1, as well as overexpression of memory-associated transcription factors including c-Jun (204). Epigenetic approaches such as low-dose decitabine priming during manufacturing have similarly shown promise by promoting memory-like phenotypes and enhancing long-term antitumor activity (205). Together, these modifications seek to generate CAR products capable of sustained activity within chronically antigen-exposed solid tumor environments.
6.3. Enhancing trafficking and regional delivery
Insufficient trafficking and poor tumor infiltration remain major limitations of CAR therapies in solid tumors. Engineering CAR cells to express chemokine receptors that match the chemokine profile of GI tumors has emerged as a promising strategy. Overexpression of receptors such as CCR2, CXCR2, and CXCR6 enhances migration toward tumor-derived chemokines and improves intratumoral accumulation in preclinical models (206, 207).
Physical barriers created by dense extracellular matrix and desmoplastic stroma further restrict immune-cell penetration, particularly in PDAC. Heparanase-expressing CAR cells have been developed to degrade extracellular matrix components and facilitate deeper tumor infiltration.
Regional delivery strategies offer an alternative approach bypassing systemic trafficking barriers. Intraperitoneal administration has shown promise for peritoneal metastases. In a phase I trial of hypoxia-responsive CEA-targeted CAR-T cells, intraperitoneal delivery produced higher response rates than intravenous administration, particularly among patients with peritoneal disease and high CEA expression (61). Similarly, hepatic artery infusion has demonstrated feasibility and evidence of CAR-T cell infiltration in colorectal liver metastases (170). These findings support the growing role of locoregional CAR delivery as a means of improving efficacy while limiting systemic toxicity.
6.4. Allogeneic and off-the-shelf platforms
The logistical challenges of autologous CAR-T manufacturing including prolonged production times, manufacturing failures, and high costs have driven interest in allogeneic platforms. CAR-NK cells are being actively investigated as an off-the-shelf cellular therapy platform, although their clinical development in GI malignancies remains at an early stage. Because NK cells function independently of HLA matching and do not induce GvHD, they offer important safety and manufacturing advantages over conventional CAR-T cells (177, 203).
Multiple CAR-NK sources are being explored, including cord blood, peripheral blood, NK-92 cell lines, and induced pluripotent stem cells (iPSCs). Among these, iPSC-derived CAR-NK cells are particularly attractive because they allow standardized large-scale manufacturing and extensive genetic engineering before differentiation into mature effector cells (208). Clinical proof-of-concept for this platform has been demonstrated outside GI oncology by Ghobadi et al. (2025), who reported the feasibility and safety of iPSC-derived CD19-directed CAR-NK cells in patients with B-cell lymphoma in a first-in-human phase I trial (200). Cytokine armoring, especially with IL-15, has further improved the persistence and activity of these products.
Allogeneic CAR-T cells are also advancing through increasingly sophisticated gene-editing strategies. Products incorporating TCR disruption, HLA editing, and resistance to inhibitory signaling pathways are entering clinical evaluation. CTX131, a CRISPR-engineered CD70-directed allogeneic CAR-T product currently being evaluated in pancreatic and esophageal cancers, exemplifies this next generation of highly engineered cellular therapies.
6.5. Future Outlook
The future of CAR-based therapy in GI oncology will likely depend on combining multiple engineering innovations rather than relying on any single modification. Multi-antigen targeting, logic-gated activation systems, armored CAR constructs, regional delivery approaches, and off-the-shelf cellular platforms are increasingly being integrated into unified therapeutic designs. Advances in synthetic biology and gene editing are enabling the development of programmable immune cells capable of sensing and adapting to the complex GI TME (181, 209).
Emerging technologies such as in vivo CAR programming using lipid nanoparticles or viral vectors may ultimately eliminate the need for ex vivo manufacturing altogether, offering a potentially scalable and cost-effective alternative to current approaches (210, 211). Preclinical proof-of-concept has been demonstrated for FAP-directed in vivo CAR-T generation using targeted lipid nanoparticles in pancreatic cancer models (212). As these innovations continue to mature, the field is moving beyond conventional CAR-T therapy toward increasingly sophisticated cellular IO platforms capable of addressing the unique biological challenges of GI malignancies.
6.6. Limitations of this review
Several limitations of this review should be acknowledged. First, this is a narrative rather than a systematic review, and the literature search was not conducted according to PRISMA guidelines; therefore, relevant studies may have been inadvertently omitted. Second, several data points cited in this review are derived from conference abstracts (ASCO 2025 and 2026) that have not yet undergone full peer review, and the reported findings may change upon final publication. Third, the clinical evidence base for CAR-based therapies in GI malignancies remains dominated by small, single-arm, phase I studies, most of which were conducted at specialized academic centers, predominantly in China and the United States. This limits the generalizability of the findings and introduces potential publication bias toward positive results. Fourth, head-to-head comparisons between different CAR targets, CAR constructs, or CAR-T versus CAR-NK platforms are lacking, making it difficult to draw definitive conclusions about the relative merits of different approaches. Fifth, the rapidly evolving nature of the field means that new data may have emerged between the completion of this review and its publication. Despite these limitations, this review provides a scoping synthesis of the current landscape and identifies key priorities for future investigation.
7. Conclusion
CAR-based cellular therapies hold significant promise for GI oncology, yet translating the success of CAR-T cells in hematologic malignancies to GI solid tumors has proved substantially more difficult. The barriers are well documented: antigen heterogeneity, poor tumor trafficking, dense desmoplastic stroma, an immunosuppressive TME, treatment-related toxicities, and manufacturing constraints.
Nevertheless, the past decade has witnessed measurable progress. Among the six clinically advanced targets reviewed here; CLDN18.2, GUCY2C, mesothelin, GPC3, CEA, and HER2 CLDN18.2-directed CAR-T therapy currently offers the strongest clinical evidence, representing the first CAR-based approach to demonstrate randomized clinical benefit in a solid tumor. CAR-NK cells represent an emerging complementary platform with potential advantages related to allogeneic manufacturing and toxicity; however, their clinical evidence in GI malignancies remains limited and substantially less mature than that of CAR-T therapy.
Advances in cellular engineering are rapidly expanding the therapeutic landscape. Armored CARs, multi-antigen targeting, logic-gated circuits, checkpoint-resistant constructs, chemokine receptor engineering, regional delivery, and allogeneic platforms are all under active development. These innovations are transforming CAR therapies from single-antigen recognition systems into programmable cellular platforms capable of dynamic interaction with the tumor microenvironment.
Achieving durable clinical benefit will require integration of multiple complementary strategies rather than optimization of any single component. Rational target selection, biomarker-guided patient stratification TME modulation, improved manufacturing, and combination approaches with existing systemic therapies will all be essential. Persistent challenges remain, but the rapid pace of innovation suggests that CAR-T therapies may become increasingly integrated into future treatment paradigms for GI malignancies, while CAR-NK approaches remain an earlier-stage investigational platform requiring further clinical validation. Realizing this potential will demand continued interdisciplinary collaboration among translational scientists, bioengineers, and clinical investigators, along with rigorous trial designs and a commitment to learning from both successes and failures.
Acknowledgments
The authors thank their respective institutions for their support during the preparation of this manuscript.
Funding Statement
The author(s) declared that financial support was not received for this work and/or its publication.
Footnotes
Edited by: Zsolt Kovács, Sciences and Technology of Târgu Mureş, Romania
Reviewed by: Zhengyang Sun, Sidney Kimmel Medical College (SKMC), United States
Elena V. Abakushina, Tecon Medical Devices LLC, Russia
Author contributions
MI: Formal analysis, Conceptualization, Investigation, Data curation, Software, Methodology, Validation, Visualization, Writing – original draft, Resources, Funding acquisition, Writing – review & editing. NA-G: Writing – original draft, Writing – review & editing. EA-N: Project administration, Visualization, Methodology, Formal analysis, Writing – original draft, Conceptualization, Writing – review & editing, Software, Data curation, Funding acquisition, Validation, Supervision, Resources, Investigation. ZA: Writing – original draft, Writing – review & editing. NMu: Writing – review & editing, Writing – original draft. NMa: Writing – review & editing, Writing – original draft. YH: Writing – original draft, Writing – review & editing. SH: Writing – review & editing, Writing – original draft. AE: Funding acquisition, Writing – original draft, Formal analysis, Resources, Software, Visualization, Project administration, Supervision, Methodology, Writing – review & editing, Validation, Conceptualization, Data curation, Investigation.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author AE declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
- 1. Danpanichkul P, Pang Y, Tothanarungroj P, Dejvajara D, Kim D, Saokhieo P, et al. Gastrointestinal cancer statistics in 2022 and projection to 2050: GLOBOCAN estimates across 185 countries. (2026) 132. [DOI] [PubMed] [Google Scholar]
- 2. Ismail M, Al-Gizey N, Al-Najjar E, Alabed Z, Mustafa N, Hamdaneh Y, et al. Bispecific antibodies in gastrointestinal cancers: current applications and future perspectives. Front Oncol. (2026) 16. doi: 10.3389/fonc.2026.1927829 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Ismail M, Alabed Z, Alhallaq K, Al-Najjar E, Mustafa N, Aldroubi Y, et al. Next-generation immune checkpoint inhibitors in gastrointestinal cancers: mechanisms, resistance, and emerging therapeutic strategies. Cancers. (2026) 18:2593. doi: 10.3390/cancers18162593 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Abdelrahim M, Esmail A, Al-Judaibi B, Khorshid O, Sherrie B, Mazzaferro V. Peri-transplant use of immune checkpoint inhibitors in hepatocellular carcinoma: a transplant oncology perspective on safety, timing, post-transplant management, and future directions. Front Oncol. (2026) 16. doi: 10.3389/fonc.2026.1826765 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Abdelrahim W, Al-Najjar E, El Beheary S, Esmail A. Prolonged disease stability with durvalumab and tremelimumab rechallenge following prior immune checkpoint inhibitors: a case report. Oncol Res. (2026) 34:28. doi: 10.32604/or.2026.083193 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Abdulgader L, Esmail A. Advances in systemic therapy for unresectable hepatocellular carcinoma: commentary on the impact of the STRIDE regimen in HIMALAYA trial. Oncol Res. (2026) 34:28. doi: 10.32604/or.2026.069227 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Abdelrahim M, Esmail A, Al-Judaibi B. Durvalumab and bevacizumab plus transarterial chemoembolization in unresectable hepatocellular carcinoma: a new paradigm? Transl Gastroenterol Hepatol. (2026) 11:1. doi: 10.21037/tgh-25-34 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Abdelrahim M, Esmail A. The use of durvalumab and tremelimumab after atezolizumab and bevacizumab in patients with hepatocellular carcinoma: case report and literature review. Immunotherapy. (2025) 17:783–90. doi: 10.1080/1750743x.2025.2539064 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Abdelrahim M, Esmail A, Kim RD, Arora SP, Arshad J, Kournoutas IA, et al. Real-world experiences using atezolizumab + bevacizumab for the treatment of unresectable hepatocellular carcinoma: a multicenter study. Cancers. (2025) 17:1814. doi: 10.3390/cancers17111814 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Abdelrahim M, Esmail A, He AR, Khushman M, Rayyan Y. Advances in immunotherapy for transplant oncology. Cancers. (2024) 16:2369. doi: 10.3390/cancers16132369 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Esmail A, Rayyan Y, Rayyan H, Abdelrahim M. Evaluating standardization of neoadjuvant immunotherapy in transplant-eligible hepatocellular carcinoma populations. Jordan Med J. (2025) 59(1):159–85. doi: 10.35516/jmj.v59i1.3957 [DOI] [Google Scholar]
- 12. Janjigian YY, Shitara K, Moehler M, Garrido M, Salman P, Shen L, et al. First-line nivolumab plus chemotherapy versus chemotherapy alone for advanced gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma (CheckMate 649): a randomised, open-label, phase 3 trial. Lancet. (2021) 398:27–40. doi: 10.1016/s0140-6736(21)00797-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Stoop TF, Javed AA, Oba A, Koerkamp BG, Seufferlein T, Wilmink JW, et al. Pancreatic cancer. Lancet. (2025) 405:1182–202. doi: 10.1016/s0140-6736(25)00261-2 [DOI] [PubMed] [Google Scholar]
- 14. André T, Elez E, Lenz H-J, Jensen LH, Touchefeu Y, Van Cutsem E, et al. Nivolumab plus ipilimumab versus nivolumab in microsatellite instability-high metastatic colorectal cancer (CheckMate 8HW): a randomised, open-label, phase 3 trial. (2025) 405:383–95. [DOI] [PubMed] [Google Scholar]
- 15. Chao J, Fuchs CS, Shitara K, Tabernero J, Muro K, Van Cutsem E, et al. Assessment of pembrolizumab therapy for the treatment of microsatellite instability–high gastric or gastroesophageal junction cancer among patients in the KEYNOTE-059, KEYNOTE-061, and KEYNOTE-062 clinical trials. (2021) 7:895–902. doi: 10.1001/jamaoncol.2021.0275 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Rha SY, Oh D-Y, Yañez P, Bai Y, Ryu M-H, Lee J, et al. Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for HER2-negative advanced gastric cancer (KEYNOTE-859): a multicentre, randomised, double-blind, phase 3 trial. (2023) 24:1181–95. doi: 10.1016/s1470-2045(23)00515-6 [DOI] [PubMed] [Google Scholar]
- 17. Haque E, Muhsen IN, Esmail A, Umoru G, Mylavarapu C, Ajewole VB, et al. Case report: Efficacy and safety of regorafenib plus fluorouracil combination therapy in the treatment of refractory metastatic colorectal cancer. Front Oncol. (2022) 12. doi: 10.3389/fonc.2022.992455 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Abdelrahim M, Esmail A, Hibi T, Mazzaferro V. Focal update on immunotherapy and liver transplantation in the era of transplant oncology. Curr Oncol. (2024) 31:5021–6. doi: 10.3390/curroncol31090371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Al-Najjar E, Esmail A, Khasawneh B, Abdelrahim S, Abdelrahim M. Idiopathic duodenal hematoma: a case report and literature review. Reports. (2025) 8:73. doi: 10.3390/reports8020073 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Esmail A, Xu J, Burns EA, Abboud K, Sheikh A, Umoru G, et al. The impact of infections in patients treated with atezolizumab plus bevacizumab for unresectable hepatocellular carcinoma. J Clin Med. (2024) 13:4994. doi: 10.3390/jcm13174994 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Esmail A, Dhillon V, Al-Najjar E, Khasawneh B, Alghamdi M, Ibnshamsah F, et al. Impact of tumor location on survival outcomes in pancreatic head versus body/tail cancer: institutional experience. Cancers. (2025) 17:1777. doi: 10.3390/cancers17111777 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Esmail A, Hassanain H, Xiu J, Abdelrahim M. The role of AI-powered molecular profiling in the diagnosis and management of cancers of unknown primary: a case report and literature review. Front Oncol. (2026) 16. doi: 10.3389/fonc.2026.1664264 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Abdelrahim S, Al-Najjar E, Esmail A. Unusual dual metastasis of hepatocellular carcinoma to the orbit and oral cavity: a case report and literature review. Explor Med. (2026) 7:1001401. doi: 10.37349/emed.2026.1001401 [DOI] [Google Scholar]
- 24. Brudno JN, Maus MV, Hinrichs CSJJ. CAR T cells and T-cell therapies for cancer: a translational science review. (2024) 332:1924–35. doi: 10.1001/jama.2024.19462 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Cappell KM, Kochenderfer JNJNRCO. Long-term outcomes following CAR T cell therapy: what we know so far. (2023) 20:359–71. doi: 10.1038/s41571-023-00754-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Abreu T, Godinho-Santos A, Amaral AT, Paiva A, Moreira JN, Goncalves JJJJOCR. Next generation CAR-T cells to tackle down solid tumors. (2026) 391:114650. doi: 10.1016/j.jconrel.2026.114650 [DOI] [PubMed] [Google Scholar]
- 27. Lamplugh ZL, Wellhausen N, June CH, Fan YJNRI. Microenvironmental regulation of solid tumour resistance to CAR T cell therapy. (2025) 26(3):230–48. doi: 10.1038/s41577-025-01229-3 [DOI] [PubMed] [Google Scholar]
- 28. Kronig MN, Wehrli M, Salas‐Benito D, Maus MVJIR. Hurdles race for CAR T‐cell therapy in digestive tract cancer. (2023) 320:100–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Qi C, Liu C, Peng Z, Zhang Y, Wei J, Qiu W, et al. Claudin-18 isoform 2-specific CAR T-cell therapy (satri-cel) versus treatment of physician’s choice for previously treated advanced gastric or gastro-oesophageal junction cancer (CT041-ST-01): a randomised, open-label, phase 2 trial. (2025) 405:2049–60. doi: 10.1016/s0140-6736(25)00860-8 [DOI] [PubMed] [Google Scholar]
- 30. Liu E, Marin D, Banerjee P, Macapinlac HA, Thompson P, Basar R, et al. Use of CAR-transduced natural killer cells in CD19-positive lymphoid tumors. (2020) 382:545–53. doi: 10.1056/nejmoa1910607 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Dagher OK, Posey JA. Forks in the road for CAR T and CAR NK cell cancer therapies. (2023) 24:1994–2007. doi: 10.1038/s41590-023-01659-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Li Y, Rezvani K, Rafei HJIR. Next‐generation chimeric antigen receptors for T‐and natural killer‐cell therapies against cancer. (2023) 320:217–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Gross G, Waks T, Eshhar ZJPOTNAOS. Expression of immunoglobulin-T-cell receptor chimeric molecules as functional receptors with antibody-type specificity. (1989) 86:10024–8. doi: 10.1073/pnas.86.24.10024 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Eshhar Z, Waks T, Gross G, Schindler DGJPOTNAOS. Specific activation and targeting of cytotoxic lymphocytes through chimeric single chains consisting of antibody-binding domains and the gamma or zeta subunits of the immunoglobulin and T-cell receptors. (1993) 90:720–4. doi: 10.1073/pnas.90.2.720 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Van Der Stegen SJ, Hamieh M, Sadelain MJNRDD. The pharmacology of second-generation chimeric antigen receptors. (2015) 14:499–509. doi: 10.1038/nrd4597 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Chmielewski M, Abken HJEOOBT. TRUCKs: the fourth generation of CARs. (2015) 15:1145–54. doi: 10.1517/14712598.2015.1046430 [DOI] [PubMed] [Google Scholar]
- 37. Kagoya Y, Tanaka S, Guo T, Anczurowski M, Wang C-H, Saso K, et al. A novel chimeric antigen receptor containing a JAK–STAT signaling domain mediates superior antitumor effects. (2018) 24:352–9. doi: 10.1038/nm.4478 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Maude SL, Laetsch TW, Buechner J, Rives S, Boyer M, Bittencourt H, et al. Tisagenlecleucel in children and young adults with B-cell lymphoblastic leukemia. (2018) 378:439–48. doi: 10.1056/nejmoa1709866 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Neelapu SS, Locke FL, Bartlett NL, Lekakis LJ, Miklos DB, Jacobson CA, et al. Axicabtagene ciloleucel CAR T-cell therapy in refractory large B-cell lymphoma. (2017) 377:2531–44. doi: 10.1056/nejmoa1707447 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Rodriguez-Otero P, Ailawadhi S, Arnulf B, Patel K, Cavo M, Nooka AK, et al. Ide-cel or standard regimens in relapsed and refractory multiple myeloma. (2023) 388:1002–14. doi: 10.1056/nejmoa2213614 [DOI] [PubMed] [Google Scholar]
- 41. Bix M, Liao N-S, Zijlstra M, Loring J, Jaenisch R, Raulet DJN. Rejection of class I MHC-deficient haemopoietic cells by irradiated MHC-matched mice. (1991) 349:329–31. doi: 10.1038/349329a0 [DOI] [PubMed] [Google Scholar]
- 42. Bauer S, Groh V, Wu J, Steinle A, Phillips JH, Lanier LL, et al. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA. (1999) 285:727–32. doi: 10.1126/science.285.5428.727 [DOI] [PubMed] [Google Scholar]
- 43. Prager I, Liesche C, Van Ooijen H, Urlaub D, Verron Q, Sandström N, et al. NK cells switch from granzyme B to death receptor–mediated cytotoxicity during serial killing. (2019) 216:2113–27. doi: 10.1084/jem.20181454 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Marin D, Li Y, Basar R, Rafei H, Daher M, Dou J, et al. Safety, efficacy and determinants of response of allogeneic CD19-specific CAR-NK cells in CD19+ B cell tumors: a phase 1/2 trial. (2024) 30:772–84. doi: 10.1038/s41591-023-02785-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Lei W, Liu H, Deng W, Chen W, Liang Y, Gao W, et al. Safety and feasibility of 4-1BB co-stimulated CD19-specific CAR-NK cell therapy in refractory/relapsed large B cell lymphoma: a phase 1 trial. (2025) 6:786–800. doi: 10.1038/s43018-025-00940-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Agliardi G, Dias J, Rampotas A, Garcia J, Roddie C. Accelerating and optimising CAR T-cell manufacture to deliver better patient products. Lancet Haematology. (2025) 12:e57–67. doi: 10.1016/s2352-3026(24)00273-4 [DOI] [PubMed] [Google Scholar]
- 47. Abken H. CAR T cell therapies in gastrointestinal cancers: current clinical trials and strategies to overcome challenges. Nat Rev Gastroenterol Hepatol. (2025) 22:463–80. doi: 10.1038/s41575-025-01062-y [DOI] [PubMed] [Google Scholar]
- 48. Bagley SJ, O’Rourke DM. Clinical investigation of CAR T cells for solid tumors: lessons learned and future directions. Pharmacol Ther. (2020) 205:107419. doi: 10.1016/j.pharmthera.2019.107419 [DOI] [PubMed] [Google Scholar]
- 49. Daei Sorkhabi A, Mohamed Khosroshahi L, Sarkesh A, Mardi A, Aghebati-Maleki A, Aghebati-Maleki L, et al. The current landscape of CAR T-cell therapy for solid tumors: mechanisms, research progress, challenges, and counterstrategies. Front Immunol. (2023) 14:1113882. doi: 10.3389/fimmu.2023.1113882 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Lu Y, Zhao F. Strategies to overcome tumour relapse caused by antigen escape after CAR T therapy. Mol Cancer. (2025) 24:126. doi: 10.1186/s12943-025-02334-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Wang D, Li B, Shen G, Zhang H, Gao Y, Du Z, et al. NKG2D CAR-NK adoptive cellular immunotherapy combined with or without PD-1 blockade in the treatment of patients with metastatic colorectal cancer: an exploratory study. (2025) 74:1–11. doi: 10.1007/s00262-025-04196-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Barrett AM, Britton ZT, Carrasco RA, Breen S, Broggi MA, Hatke AL, et al. Preclinical evaluation of AZD6422, an armored chimeric antigen receptor T cell targeting CLDN18. 2 in gastric, pancreatic, and esophageal cancers. (2024) 30:5413–29. doi: 10.1158/1078-0432.ccr-24-1853 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Qi C, Gong J, Li J, Liu D, Qin Y, Ge S, et al. Claudin18. 2-specific CAR T cells in gastrointestinal cancers: phase 1 trial interim results. (2022) 28:1189–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Steffin D, Ghatwai N, Montalbano A, Rathi P, Courtney AN, Arnett AB, et al. Interleukin-15-armoured GPC3 CAR T cells for patients with solid cancers. (2025) 637:940–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Chen N, Pu C, Zhao L, Li W, Wang C, Zhu R, et al. Chimeric antigen receptor T cells targeting CD19 and GCC in metastatic colorectal cancer: A nonrandomized clinical trial. JAMA Oncol. (2024) 10:1532–6. doi: 10.1001/jamaoncol.2024.3891 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Barber-Rotenberg JS, Haas AR, Aggarwal C, O’Hara M, Hexner E, Pequignot E, et al. Phase 1 study of autologous T cells bearing fully human chimeric antigen receptors targeting mesothelin in mesothelin-expressing cancers. Mol Ther J Am Soc Gene Ther. (2026) 34:2653–65. doi: 10.1016/j.ymthe.2026.01.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57. Beatty GL, O’Hara MH, Lacey SF, Torigian DA, Nazimuddin F, Chen F, et al. Activity of mesothelin-specific chimeric antigen receptor T cells against pancreatic carcinoma metastases in a phase 1 trial. (2018) 155:29–32. doi: 10.1053/j.gastro.2018.03.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58. Haas AR, Tanyi JL, O’Hara MH, Gladney WL, Lacey SF, Torigian DA, et al. Phase I study of lentiviral-transduced chimeric antigen receptor-modified T cells recognizing mesothelin in advanced solid cancers. Mol Ther J Am Soc Gene Ther. (2019) 27:1919–29. doi: 10.1016/j.ymthe.2019.07.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Hassan R, Butler M, O’Cearbhaill RE, Oh DY, Johnson M, Zikaras K, et al. Mesothelin-targeting T cell receptor fusion construct cell therapy in refractory solid tumors: phase 1/2 trial interim results. (2023) 29:2099–109. doi: 10.1038/s41591-023-02452-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60. Liu X, Onda M, Schlomer J, Bassel L, Kozlov S, Tai C-H, et al. Tumor resistance to anti-mesothelin CAR-T cells caused by binding to shed mesothelin is overcome by targeting a juxtamembrane epitope. (2024) 121:. doi: 10.1073/pnas.2317283121 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61. Gao Y, Li J, Zhang H, Zhang Y, Qian S, Zhu X, et al. Hypoxia-responsive CEA-targeted CAR T cells in CEA-positive solid tumors through intraperitoneal or intravenous infusion: a phase 1 trial. (2026) 7(4):608–21. doi: 10.1038/s43018-026-01124-3 [DOI] [PubMed] [Google Scholar]
- 62. Zhang B-L, Li D, Gong Y-L, Huang Y, Qin D-Y, Jiang L, et al. Preclinical evaluation of chimeric antigen receptor–modified T cells specific to epithelial cell adhesion molecule for treating colorectal cancer. (2019) 30:402–12. doi: 10.1089/hum.2018.229 [DOI] [PubMed] [Google Scholar]
- 63. Sandberg ML, Wang X, Martin AD, Nampe DP, Gabrelow GB, Li CZ, et al. A carcinoembryonic antigen-specific cell therapy selectively targets tumor cells with HLA loss of heterozygosity in vitro and in vivo. Sci Transl Med. (2022) 14:eabm0306. doi: 10.1126/scitranslmed.abm0306 [DOI] [PubMed] [Google Scholar]
- 64. Shi D, Shi Y, Kaseb AO, Qi X, Zhang Y, Chi J, et al. Chimeric antigen receptor-glypican-3 T-cell therapy for advanced hepatocellular carcinoma: results of phase I trials. (2020) 26:3979–89. doi: 10.1158/1078-0432.ccr-19-3259 [DOI] [PubMed] [Google Scholar]
- 65. Sun H, Xing C, Jiang S, Yu K, Dai S, Kong H, et al. Long term complete response of advanced hepatocellular carcinoma to glypican-3 specific chimeric antigen receptor T-Cells plus sorafenib, a case report. Front Immunol. (2022) 13:963031. doi: 10.3389/fimmu.2022.963031 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66. Yang L, Pham K, Xi Y, Wu Q, Liu D, Robertson KD, et al. Exploring Glypican-3 targeted CAR-NK treatment and potential therapy resistance in hepatocellular carcinoma. (2025) 20:. doi: 10.1371/journal.pone.0317401 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67. Xu J, Meng Q, Sun H, Zhang X, Yun J, Li B, et al. HER2-specific chimeric antigen receptor-T cells for targeted therapy of metastatic colorectal cancer. (2021) 12:1109. doi: 10.1038/s41419-021-04100-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68. Niwa Y, Matsushita S, Nakayama I, Kondoh M. Targeting claudins in cancer. Nat Rev Drug Discov. (2026) 25(9):723–38. doi: 10.1038/s41573-026-01450-2 [DOI] [PubMed] [Google Scholar]
- 69. Saviano A, Toso A, Klempner SJ, Baumert TF. Claudin proteins as emerging therapeutic targets for solid tumours. Nat Rev Cancer. (2026) 26:364–83. doi: 10.1038/s41568-026-00913-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70. Nakayama I, Qi C, Chen Y, Nakamura Y, Shen L, Shitara K. Claudin 18.2 as a novel therapeutic target. Nat Rev Clin Oncol. (2024) 21:354–69. doi: 10.1038/s41571-024-00874-2 [DOI] [PubMed] [Google Scholar]
- 71. Grizzi G, Venetis K, Denaro N, Bonomi M, Celotti A, Pagkali A, et al. Anti-claudin treatments in gastroesophageal adenocarcinoma: mainstream and upcoming strategies. J Clin Med. (2023) 12(8):2973. doi: 10.3390/jcm12082973 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 72. Hashimoto T, Iida N, Nakamura Y, Nonomura N, Morizane C, Iwata H, et al. Landscape analysis of CLDN18 expression and isoform distribution in solid tumors: insights from MONSTAR-SCREEN-2 study. Cancer Sci. (2025) 116:2218–31. doi: 10.1111/cas.70100 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 73. Al Subhi M, Al-Nabhani D, Al Sarmi M, Abd Elrahman A, AlSawafi Y, Sirasanagandla SR, et al. Expression of Claudin-18.2 in pancreatic ductal adenocarcinoma: a systematic review and meta-analysis. (2026) 13:1809374. doi: 10.3389/fmed.2026.1809374 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74. McHugh KE, Pai RK, Grant RC, Gallinger S, Davison J, Ma C, et al. Claudin 18.2 expression in 1404 digestive tract adenocarcinomas including 1175 colorectal carcinomas: distinct colorectal carcinoma subtypes are claudin 18.2 positive. Modern Pathol Off J United States Can Acad Pathology Inc. (2025) 38:100712. doi: 10.1016/j.modpat.2025.100712 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 75. Jiang H, Shi Z, Wang P, Wang C, Yang L, Du G, et al. Claudin18.2-specific chimeric antigen receptor engineered T cells for the treatment of gastric cancer. J Natl Cancer Institute. (2019) 111:409–18. doi: 10.1093/jnci/djy134 [DOI] [PubMed] [Google Scholar]
- 76. Shi H, Li A, Dai Z, Xue J, Zhao Q, Tian J, et al. IL-15 armoring enhances the antitumor efficacy of claudin 18.2-targeting CAR-T cells in syngeneic mouse tumor models. Front Immunol. (2023) 14:1165404. doi: 10.3389/fimmu.2023.1165404 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 77. Qi C, Liu C, Gong J, Liu D, Wang X, Zhang P, et al. Claudin18.2-specific CAR T cells in gastrointestinal cancers: phase 1 trial final results. Nat Med. (2024) 30:2224–34. doi: 10.1038/s41591-024-03037-z [DOI] [PubMed] [Google Scholar]
- 78. Botta GP, Kelly RJ, Jin Z, Ma H, Ku GY, Zhao D, et al. CLDN18.2 chimeric antigen receptor T cell therapy for patients with advanced gastric and pancreatic adenocarcinoma: Results of ELIMYN18.2 phase 1b clinical trial. (2024) 42:356. doi: 10.1200/jco.2024.42.3_suppl.356 [DOI] [Google Scholar]
- 79. Klempner SJ, Lee K-W, Shitara K, Metges J-P, Lonardi S, Ilson DH, et al. ILUSTRO: phase II multicohort trial of zolbetuximab in patients with advanced or metastatic claudin 18.2–positive gastric or gastroesophageal junction adenocarcinoma. (2023) 29:3882–91. doi: 10.1158/1078-0432.ccr-23-0204 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 80. Keam SJJD. Zolbetuximab: first approval: SJ keam. (2024) 84:977–83. [DOI] [PubMed] [Google Scholar]
- 81. Magee MS, Kraft CL, Abraham TS, Baybutt TR, Marszalowicz GP, Li P, et al. GUCY2C-directed CAR-T cells oppose colorectal cancer metastases without autoimmunity. Oncoimmunology. (2016) 5:e1227897. doi: 10.1080/2162402x.2016.1227897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 82. Wilson C, Lin JE, Li P, Snook AE, Gong J, Sato T, et al. The paracrine hormone for the GUCY2C tumor suppressor, guanylin, is universally lost in colorectal cancer. Cancer Epidemiology Biomarkers Prev A Publ Am Assoc For Cancer Research cosponsored by Am Soc Prev Oncol. (2014) 23:2328–37. doi: 10.1158/1055-9965.epi-14-0440 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 83. Sundar R, Nakayama I, Markar SR, Shitara K, van Laarhoven HW, Janjigian YY, et al. Gastric cancer. (2025) 405:2087–102. doi: 10.1016/s0140-6736(25)00052-2 [DOI] [PubMed] [Google Scholar]
- 84. Shah MA, Shitara K, Ajani JA, Bang YJ, Enzinger P, Ilson D, et al. Zolbetuximab plus CAPOX in CLDN18.2-positive gastric or gastroesophageal junction adenocarcinoma: the randomized, phase 3 GLOW trial. Nat Med. (2023) 29:2133–41. doi: 10.1038/s41591-023-02465-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 85. Shitara K, Lordick F, Bang YJ, Enzinger P, Ilson D, Shah MA, et al. Zolbetuximab plus mFOLFOX6 in patients with CLDN18.2-positive, HER2-negative, untreated, locally advanced unresectable or metastatic gastric or gastro-oesophageal junction adenocarcinoma (SPOTLIGHT): a multicentre, randomised, double-blind, phase 3 trial. Lancet (London England). (2023) 401:1655–68. doi: 10.1016/s0140-6736(23)00620-7 [DOI] [PubMed] [Google Scholar]
- 86. Liu C, Qi C, Gong J, Liu D, Zhang P, Qin Y, et al. Long-term follow-up of satricabtagene autoleucel (satri-cel) as sequential therapy after first-line treatment for advanced gastric cancer: A subgroup analysis. (2026) 44:2557. doi: 10.1200/jco.2026.44.16_suppl.2557 [DOI] [Google Scholar]
- 87. Carrithers SL, Barber MT, Biswas S, Parkinson SJ, Park PK, Goldstein SD, et al. Guanylyl cyclase C is a selective marker for metastatic colorectal tumors in human extraintestinal tissues. PNAS. (1996) 93:14827–32. doi: 10.1073/pnas.93.25.14827 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 88. Rappaport JA, Waldman SA. The guanylate cyclase C-cGMP signaling axis opposes intestinal epithelial injury and neoplasia. Front Oncol. (2018) 8:299. doi: 10.3389/fonc.2018.00299 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 89. Blomain ES, Rappaport JA, Pattison AM, Bashir B, Caparosa E, Stem J, et al. APC-β-catenin-TCF signaling silences the intestinal guanylin-GUCY2C tumor suppressor axis. Cancer Biol Ther. (2020) 21:441–51. doi: 10.1080/15384047.2020.1721262 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 90. Magee MS, Abraham TS, Baybutt TR, Flickinger JC, Jr., Ridge NA, Marszalowicz GP, et al. Human GUCY2C-targeted chimeric antigen receptor (CAR)-expressing T cells eliminate colorectal cancer metastases. Cancer Immunol Res. (2018) 6:509–16. doi: 10.1158/2326-6066.cir-16-0362 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 91. Baybutt TR, Entezari AA, Caspi A, Staudt RE, Carlson RD, Waldman SA, et al. CD8α Structural domains enhance GUCY2C CAR-T cell efficacy. Cancer Biol Ther. (2024) 25:2398801. doi: 10.1080/15384047.2024.2398801 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 92. Xiao L, Chen N, Pu C, Zhao L, Li N, Wang C, et al. (2023). A phase 1 dose escalation study of GCC19CART: A novel coupled CAR therapy for patients with metastatic colorectal cancer. J Clin Oncol. (2023) 41:3547. doi: 10.1200/JCO.2023.41.16_suppl.35478552591 [DOI] [Google Scholar]
- 93. Keenan BP, Lieu CH, Fakih M, Venook AP, Chen D, Rowinsky EK, et al. A phase 1 dose-escalation study of GCC19CART: A novel CAR T-cell therapy for metastatic colorectal cancer in the United States. (2025) 43:175. doi: 10.1200/jco.2025.43.4_suppl.175 [DOI] [Google Scholar]
- 94. Cao M, Alvarez J, Mitra R, Xu M, Baybutt TR, Sun Z, et al. IFNγ-induced antigen loss in chimeric antigen receptor-T cell therapy. Front Immunol. (2026) 17:1772472. doi: 10.3389/fimmu.2026.1772472 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 95. Qi C, Liu C, Gong J, Wang X, Wang Z, Xu T, et al. Guanylyl cyclase 2C-targeted chimeric antigen receptor T-cell therapy in patients with metastatic colorectal cancer. J Clin Oncol Off J Am Soc Clin Oncol. (2026) 44(21):2006–16. doi: 10.1200/jco-25-01090 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 96. Chang K, Pastan I. Molecular cloning of mesothelin, a differentiation antigen present on mesothelium, mesotheliomas, and ovarian cancers. PNAS. (1996) 93:136–40. doi: 10.1073/pnas.93.1.136 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 97. Bharadwaj U, Marin-Muller C, Li M, Chen C, Yao Q. Mesothelin confers pancreatic cancer cell resistance to TNF-α-induced apoptosis through Akt/PI3K/NF-κB activation and IL-6/Mcl-1 overexpression. Mol Cancer. (2011) 10:106. doi: 10.1186/1476-4598-10-106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 98. Tang Z, Qian M, Ho M. The role of mesothelin in tumor progression and targeted therapy. Anti-Cancer Agents Med Chem. (2013) 13:276–80. doi: 10.2174/1871520611313020014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 99. Hassan R, Laszik ZG, Lerner M, Raffeld M, Postier R, Brackett D. Mesothelin is overexpressed in pancreaticobiliary adenocarcinomas but not in normal pancreas and chronic pancreatitis. Am J Clin Pathol. (2005) 124:838–45. doi: 10.1309/f1b64cl7h8vjkeaf [DOI] [PubMed] [Google Scholar]
- 100. Hecht JR, Grierson P, Welling TH, Spencer KR, Hazim AZ, Park JC, et al. Correlation of mesothelin (MSLN) expression measured by RNA sequencing (RNASeq) and immunohistochemistry (IHC) in MSLN-expressing tumors. (2025) 43:766. doi: 10.1200/jco.2025.43.4_suppl.766 [DOI] [Google Scholar]
- 101. Weidemann S, Perez D, Izbicki JR, Neipp M, Mofid H, Daniels T, et al. Mesothelin is commonly expressed in pancreatic adenocarcinoma but unrelated to cancer aggressiveness. Cancer Invest. (2021) 39:711–20. doi: 10.1080/07357907.2021.1943747 [DOI] [PubMed] [Google Scholar]
- 102. Liang O, White AL, Fielder T, Shin JS, Sagnella SM, Schmitz U, et al. High mesothelin expression is associated with low cytotoxic T cell infiltration in pancreatic cancer. Front Immunol. (2025) 16:1651687. doi: 10.3389/fimmu.2025.1651687 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 103. Duangdara J, Lin S, Hong J, Li D, Suriyonplengsaeng C, Larbcharoensub N, et al. Preclinical development of mesothelin-targeting CAR T cells for the treatment of cholangiocarcinoma. Hepatol Commun. (2026) 10(2):e0888. doi: 10.1097/hc9.0000000000000888 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 104. Yu L, Feng M, Kim H, Phung Y, Kleiner DE, Gores GJ, et al. Mesothelin as a potential therapeutic target in human cholangiocarcinoma. J Cancer. (2010) 1:141–9. doi: 10.7150/jca.1.141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 105. Kawamata F, Kamachi H, Einama T, Homma S, Tahara M, Miyazaki M, et al. Intracellular localization of mesothelin predicts patient prognosis of extrahepatic bile duct cancer. Int J Oncol. (2012) 41:2109–18. doi: 10.3892/ijo.2012.1662 [DOI] [PubMed] [Google Scholar]
- 106. Takihata Y, Einama T, Kobayashi K, Suzuki T, Yonamine N, Fujinuma I, et al. Different role of MSLN and CA125 co-expression as a prognostic predictor between perihilar and distal bile duct carcinoma. Oncol Lett. (2021) 21:414. doi: 10.3892/ol.2021.12675 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 107. Baba K, Ishigami S, Arigami T, Uenosono Y, Okumura H, Matsumoto M, et al. Mesothelin expression correlates with prolonged patient survival in gastric cancer. J Surg Oncol. (2012) 105:195–9. doi: 10.1002/jso.22024 [DOI] [PubMed] [Google Scholar]
- 108. Einama T, Homma S, Kamachi H, Kawamata F, Takahashi K, Takahashi N, et al. Luminal membrane expression of mesothelin is a prominent poor prognostic factor for gastric cancer. Br J Cancer. (2012) 107:137–42. doi: 10.1038/bjc.2012.235 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 109. Shin SJ, Park S, Kim MH, Nam CM, Kim H, Choi YY, et al. Mesothelin expression is a predictive factor for peritoneal recurrence in curatively resected stage III gastric cancer. Oncologist. (2019) 24:e1108–14. doi: 10.1634/theoncologist.2018-0896 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 110. Illei PB, Alewine C, Zahurak M, Cowan ML, Montgomery E, Hassan R, et al. Mesothelin expression in advanced gastroesophageal cancer represents a novel target for immunotherapy. Appl Immunohistochemistry Mol Morphology AIMM. (2016) 24:246–52. doi: 10.1097/pai.0000000000000292 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 111. Torchiaro E, Cortese M, Petti C, Basirico M, Invrea F, D’Andrea A, et al. Repurposing anti-mesothelin CAR-NK immunotherapy against colorectal cancer. J Transl Med. (2024) 22:1100. doi: 10.1186/s12967-024-05851-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 112. Lv J, Zhao R, Wu D, Zheng D, Wu Z, Shi J, et al. Mesothelin is a target of chimeric antigen receptor T cells for treating gastric cancer. J Hematol Oncol. (2019) 12:18. doi: 10.1186/s13045-019-0704-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 113. Tomar S, Zhang J, Khanal M, Hong J, Venugopalan A, Jiang Q, et al. Development of highly effective anti-mesothelin hYP218 chimeric antigen receptor T cells with increased tumor infiltration and persistence for treating solid tumors. Mol Cancer Ther. (2022) 21:1195–206. doi: 10.1158/1535-7163.mct-22-0073 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 114. Armstrong A, van der Plancke G, Nishiguchi S, Salas-Benito D, Bouffard AA, Goncalves S, et al. Ibrutinib and PD-1 blockade potentiate mesothelin-targeting CAR T-cell therapy in preclinical models of pancreatic cancer. Clin Cancer Res Off J Am Assoc For Cancer Res. (2026) 32:611–27. doi: 10.1158/1078-0432.CCR-25-2907 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 115. Amit U, Uslu U, Verginadis II, Kim MM, Motlagh SAO, Diffenderfer ES, et al. Proton radiation boosts the efficacy of mesothelin-targeting chimeric antigen receptor T cell therapy in pancreatic cancer. PNAS. (2024) 121:e2403002121. doi: 10.1073/pnas.2403002121 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 116. Aboalela MA, Abdelmoneim M, Matsumura S, Eissa IR, Bustos-Villalobos I, Sibal PA, et al. Enhancing mesothelin CAR T cell therapy for pancreatic cancer with an oncolytic herpes virus boosting CAR target antigen expression. Cancer Immunology Immunotherapy CII. (2025) 74:202. doi: 10.1007/s00262-025-04039-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 117. Volpe A, Zurita J, Shenker L, Mane MM, Adusumilli PS, Aleshin A, et al. Membrane-bound TRAIL-armoring augments CAR-T cells in mesothelin-positive solid malignancies. J Immunother Cancer. (2026) 14. doi: 10.1136/jitc-2025-014330 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 118. Li YR, Shen X, Zhu E, Li Z, Huang J, Le T, et al. Targeting orthotopic and metastatic pancreatic cancer with allogeneic stem cell-engineered mesothelin-redirected CAR-NKT cells. PNAS. (2025) 122:e2517786122. doi: 10.1073/pnas.2517786122 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 119. Ko AH, Jordan AC, Tooker E, Lacey SF, Chang RB, Li Y, et al. Dual targeting of mesothelin and CD19 with chimeric antigen receptor-modified T cells in patients with metastatic pancreatic cancer. Mol Ther J Am Soc Gene Ther. (2020) 28:2367–78. doi: 10.1016/j.ymthe.2020.07.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 120. Qi C, Li W, Zhang P, Liu D, Liu C, Li J, et al. Safety and preliminary efficacy of a novel armored mesothelin-targeted CAR-T therapy in advanced solid tumors. (2026) 44:2543. doi: 10.1200/jco.2026.44.16_suppl.2543 [DOI] [Google Scholar]
- 121. Spencer KR, Punekar SR, Grierson P, Mitchell J, Lin Y, Castria T, et al. EVEREST-2: A seamless phase 1/2 study of A2B694, a logic-gated Tmod chimeric antigen receptor T-cell (CAR T) therapy, in patients with pancreatic cancer (PANC) or other mesothelin (MSLN)-expressing solid tumors and human leukocyte antigen (HLA)–A*02 loss of heterozygosity (LOH). (2025) 43:TPS788. doi: 10.1200/jco.2025.43.4_suppl.tps788 [DOI] [Google Scholar]
- 122. Restle D, Amador-Molina A, Misawa K, Banerjee S, Ku G, Adusumilli PS. Intraperitoneal CAR T-cell therapy for peritoneal carcinomatosis from gastroesophageal cancer: preclinical investigations to a phase I clinical trial (NCT06623396). J Immunother Cancer. (2025) 13. doi: 10.1136/jitc-2025-012292 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 123. Capurro M, Wanless IR, Sherman M, Deboer G, Shi W, Miyoshi E, et al. Glypican-3: a novel serum and histochemical marker for hepatocellular carcinoma. (2003) 125:89–97. doi: 10.1016/s0016-5085(03)00689-9 [DOI] [PubMed] [Google Scholar]
- 124. Filmus J, Capurro M. Glypican‐3: a marker and a therapeutic target in hepatocellular carcinoma. (2013) 280:2471–6. doi: 10.1111/febs.12126 [DOI] [PubMed] [Google Scholar]
- 125. Capurro MI, Xiang Y-Y, Lobe C, Filmus J. Glypican-3 promotes the growth of hepatocellular carcinoma by stimulating canonical Wnt signaling. (2005) 65:6245–54. doi: 10.1158/0008-5472.can-04-4244 [DOI] [PubMed] [Google Scholar]
- 126. Zhou F, Shang W, Yu X, Tian J. Glypican‐3: a promising biomarker for hepatocellular carcinoma diagnosis and treatment. (2018) 38:741–67. [DOI] [PubMed] [Google Scholar]
- 127. Guo M, Zhang H, Zheng J, Liu Y. Glypican-3: a new target for diagnosis and treatment of hepatocellular carcinoma. (2020) 11:2008. doi: 10.7150/jca.39972 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 128. Batra SA, Rathi P, Guo L, Courtney AN, Fleurence J, Balzeau J, et al. Glypican-3–specific CAR T cells coexpressing IL15 and IL21 have superior expansion and antitumor activity against hepatocellular carcinoma. (2020) 8:309–20. doi: 10.1158/2326-6066.cir-19-0293 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 129. Wu X, Luo H, Shi B, Di S, Sun R, Su J, et al. Combined antitumor effects of sorafenib and GPC3-CAR T cells in mouse models of hepatocellular carcinoma. (2019) 27:1483–94. doi: 10.1016/j.ymthe.2019.04.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 130. Fu Q, Zheng Y, Fang W, Zhao Q, Zhao P, Liu L, et al. RUNX-3-expressing CAR T cells targeting glypican-3 in patients with heavily pretreated advanced hepatocellular carcinoma: a phase I trial. (2023) 63. doi: 10.1016/j.eclinm.2023.102175 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 131. Fan J, Zhou J, Huang X-W, Gao Q, Shi J, Cheng X, et al. Objective response rates with Ori-C101, an armored GPC3-directed CAR-T, in heavily pretreated advanced HCC: Updated results from the phase Ib BEACON study. (2026) 44:2508. doi: 10.1200/jco.2026.44.16_suppl.2508 [DOI] [Google Scholar]
- 132. Yu M, Luo H, Fan M, Wu X, Shi B, Di S, et al. Development of GPC3-specific chimeric antigen receptor-engineered natural killer cells for the treatment of hepatocellular carcinoma. (2018) 26:366–78. doi: 10.1016/j.ymthe.2017.12.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 133. Sun L, Gao F, Gao Z, Ao L, Li N, Ma S, et al. Shed antigen-induced blocking effect on CAR-T cells targeting Glypican-3 in Hepatocellular Carcinoma. J Immunother Cancer. (2021) 9(4):e001875. doi: 10.1136/jitc-2020-001875 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 134. Gold P, Freedman S. Demonstration of tumor-specific antigens in human colonic carcinomata by immunological tolerance and absorption techniques. (1965) 121:439–62. doi: 10.1084/jem.121.3.439 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 135. Hammarström S. (1999). “ The carcinoembryonic antigen (CEA) family: structures, suggested functions and expression in normal and malignant tissues”, in: Seminars in Cancer Biology (Elsevier: Academic Press; ). 9(2):67–81 [DOI] [PubMed] [Google Scholar]
- 136. Benchimol S, Fuks A, Jothy S, Beauchemin N, Shirota K, Stanners C. Carcinoembryonic antigen, a human tumor marker, functions as an intercellular adhesion molecule. (1989) 57:327–36. doi: 10.1016/0092-8674(89)90970-7 [DOI] [PubMed] [Google Scholar]
- 137. Bhatnagar J, Heroman W, Murphy M, Austin G. Immunohistochemical detection of carcinoembryonic antigen in esophageal carcinomas: a comparison with other gastrointestinal neoplasms. (2002) 22:1849–57. [PubMed] [Google Scholar]
- 138. Gebauer F, Wicklein D, Horst J, Sundermann P, Maar H, Streichert T, et al. Carcinoembryonic antigen-related cell adhesion molecules (CEACAM) 1, 5 and 6 as biomarkers in pancreatic cancer. (2014) 9:. doi: 10.1371/journal.pone.0113023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 139. Jansen K, Kornfeld L, Lennartz M, Dwertmann Rico S, Kind S, Reiswich V, et al. Carcinoembryonic antigen expression in human tumors: a tissue microarray study on 13,725 tumors. (2024) 16:4052. doi: 10.3390/cancers16234052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 140. Thistlethwaite FC, Gilham DE, Guest RD, Rothwell DG, Pillai M, Burt DJ, et al. The clinical efficacy of first-generation carcinoembryonic antigen (CEACAM5)-specific CAR T cells is limited by poor persistence and transient pre-conditioning-dependent respiratory toxicity. Cancer Immunology Immunotherapy CII. (2017) 66:1425–36. doi: 10.1007/s00262-017-2034-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 141. Cha SE, Kujawski M, Yazaki J, Brown C, Shively J. Tumor regression and immunity in combination therapy with anti-CEA chimeric antigen receptor T cells and anti-CEA-IL2 immunocytokine. (2021) 10:1899469. doi: 10.1080/2162402x.2021.1899469 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 142. Chi X, Yang P, Zhang E, Gu J, Xu H, Li M, et al. Significantly increased anti‐tumor activity of carcinoembryonic antigen‐specific chimeric antigen receptor T cells in combination with recombinant human IL‐12. (2019) 8:4753–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 143. Banik D, Ward CJ, Chakraborty S, Zhang Z, Heraghty D, Suresh P, et al. Enzymatic remodelling of tumour microenvironment enhances anti-CEACAM5 CAR T-cell efficacy against colorectal cancer. (2026) 13:. doi: 10.1002/advs.202509762 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 144. Katz SC, Burga RA, McCormack E, Wang LJ, Mooring W, Point GR, et al. Phase I hepatic immunotherapy for metastases study of intra-arterial chimeric antigen receptor-modified T-cell therapy for CEA+ Liver metastases. Clin Cancer Res Off J Am Assoc For Cancer Res. (2015) 21:3149–59. doi: 10.1158/1078-0432.ccr-14-1421 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 145. Katz SC, Hardaway J, Prince E, Guha P, Cunetta M, Moody A, et al. HITM-SIR: phase Ib trial of intraarterial chimeric antigen receptor T-cell therapy and selective internal radiation therapy for CEA+ liver metastases. (2020) 27:341–55. [DOI] [PubMed] [Google Scholar]
- 146. Zhang C, Wang Z, Yang Z, Wang M, Li S, Li Y, et al. Phase I escalating-dose trial of CAR-T therapy targeting CEA+ metastatic colorectal cancers. (2017) 25:1248–58. doi: 10.1016/j.ymthe.2017.03.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 147. Katz SC, Point GR, Cunetta M, Thorn M, Guha P, Espat NJ, et al. Regional CAR-T cell infusions for peritoneal carcinomatosis are superior to systemic delivery. Cancer Gene Ther. (2016) 23:142–8. doi: 10.1038/cgt.2016.14 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 148. Raghav KP, Moasser M. Molecular pathways and mechanisms of HER2 in cancer therapy. (2023) 29:2351–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 149. Press MF, Cordon-Cardo C, Slamon D. Expression of the HER-2/neu proto-oncogene in normal human adult and fetal tissues. (1990) 5:953–62. [PubMed] [Google Scholar]
- 150. Bang Y-J, Van Cutsem E, Feyereislova A, Chung HC, Shen L, Sawaki A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. (2010) 376:687–97. doi: 10.1016/s0140-6736(10)61121-x [DOI] [PubMed] [Google Scholar]
- 151. Meric-Bernstam F, Hurwitz H, Raghav KPS, McWilliams RR, Fakih M, VanderWalde A, et al. Pertuzumab plus trastuzumab for HER2-amplified metastatic colorectal cancer (MyPathway): an updated report from a multicentre, open-label, phase 2a, multiple basket study. (2019) 20:518–30. doi: 10.1016/s1470-2045(18)30904-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 152. Harding JJ, Fan J, Oh D-Y, Choi HJ, Kim JW, Chang H-M, et al. Zanidatamab for HER2-amplified, unresectable, locally advanced or metastatic biliary tract cancer (HERIZON-BTC-01): a multicentre, single-arm, phase 2b study. (2023) 24:772–82. doi: 10.1016/s1470-2045(23)00242-5 [DOI] [PubMed] [Google Scholar]
- 153. Bryant D, Feldman R, Abdulla F, Magee D, Ribeiro JR, Oberley M, et al. A real-world experience in pan-tumor testing for HER2 IHC in more than 65 000 solid tumors. (2025) 11:919–21. doi: 10.1001/jamaoncol.2025.1791 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 154. Wagner AD, Özdemir BC, Rüschoff JJ. Human epidermal growth factor receptor 2-positive digestive tumors. (2019) 31:354–61. doi: 10.1097/cco.0000000000000544 [DOI] [PubMed] [Google Scholar]
- 155. Liu X, Jiang S, Fang C, Yang S, Olalere D, Pequignot EC, et al. Affinity-tuned ErbB2 or EGFR chimeric antigen receptor T cells exhibit an increased therapeutic index against tumors in mice. (2015) 75:3596–607. doi: 10.1158/0008-5472.can-15-0159 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 156. Shabaneh TB, Stevens AR, Stull SM, Shimp KR, Seaton BW, Gad EA, et al. Systemically administered low-affinity HER2 CAR T cells mediate antitumor efficacy without toxicity. (2024) 12:. doi: 10.1136/jitc-2023-008566 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 157. Ahmed N, Brawley V, Hegde M, Bielamowicz K, Kalra M, Landi D, et al. HER2-specific chimeric antigen receptor–modified virus-specific T cells for progressive glioblastoma: a phase 1 dose-escalation trial. (2017) 3:1094–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 158. Ahmed N, Brawley VS, Hegde M, Robertson C, Ghazi A, Gerken C, et al. Human epidermal growth factor receptor 2 (HER2)–specific chimeric antigen receptor–modified T cells for the immunotherapy of HER2-positive sarcoma. (2015) 33:1688–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 159. Hosking MP, Shirinbak S, Omilusik K, Chandra S, Kaneko MK, Gentile A, et al. Preferential tumor targeting of HER2 by iPSC-derived CAR T cells engineered to overcome multiple barriers to solid tumor efficacy. (2025) 32:1087–101. doi: 10.1016/j.stem.2025.05.007 [DOI] [PubMed] [Google Scholar]
- 160. Freudenberg-Jahn K, Ben-Horin I, Murali Shankar N, Schuldt M, Ortiz-Montero P, Loureiro LR, et al. Dual CAR-NK cells targeting PD-L1 and ErbB2 (HER2) exhibit cooperative CAR signaling and counteract solid tumor heterogeneity. (2026) 45(1):127. doi: 10.1186/s13046-026-03722-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 161. Schönfeld K, Sahm C, Zhang C, Naundorf S, Brendel C, Odendahl M, et al. Selective inhibition of tumor growth by clonal NK cells expressing an ErbB2/HER2-specific chimeric antigen receptor. (2015) 23:330–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 162. Morgan RA, Yang JC, Kitano M, Dudley ME, Laurencot CM, Rosenberg SA. Case report of a serious adverse event following the administration of T cells transduced with a chimeric antigen receptor recognizing ERBB2. (2010) 18:843–51. doi: 10.1038/mt.2010.24 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 163. Feng K, Liu Y, Guo Y, Qiu J, Wu Z, Dai H, et al. Phase I study of chimeric antigen receptor modified T cells in treating HER2-positive advanced biliary tract cancers and pancreatic cancers. Protein Cell. (2018) 9:838–47. doi: 10.1007/s13238-017-0440-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 164. Dumbrava EE, Olson D, Gouda MA, George MA, Saibil SD, Apostolopoulou M, et al. Autologous T cell antigen coupler targeting HER2 (TAC01-HER2) in advanced or metastatic solid tumors. (2026) 37(10):1358–66. doi: 10.1016/j.annonc.2026.05.696 [DOI] [PubMed] [Google Scholar]
- 165. Burger MC, Forster M-T, Romanski A, Straßheimer F, Macas J, Zeiner PS, et al. Intracranial injection of natural killer cells engineered with a HER2-targeted chimeric antigen receptor in patients with recurrent glioblastoma. (2023) 25:2058–71. doi: 10.1093/neuonc/noad087 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 166. Schlechter BL, Olson D, George MA, Saibil S, Giordano A, Bouvier R, et al. A phase I/II trial investigating safety and efficacy of autologous TAC01-HER2 in relapsed or refractory solid tumors. (2023) 41:2519. doi: 10.1158/1535-7163.targ-23-b035 [DOI] [Google Scholar]
- 167. Entezari AA, Snook AE, Waldman SA. Guanylyl cyclase 2C (GUCY2C) in gastrointestinal cancers: recent innovations and therapeutic potential. (2021) 25:335–46. doi: 10.1080/14728222.2021.1937124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 168. Snook AE, Magee MS, Waldman SA. GUCY2C-targeted cancer immunotherapy: past, present and future. (2011) 51:161–9. doi: 10.1007/s12026-011-8253-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 169. Flugel CL, Majzner RG, Krenciute G, Dotti G, Riddell SR, Wagner DL, et al. Overcoming on-target, off-tumour toxicity of CAR T cell therapy for solid tumours. Nat Rev Clin Oncol. (2023) 20:49–62. doi: 10.1038/s41571-022-00704-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 170. Kronig MN, Wehrli M, Salas-Benito D, Maus MV. Hurdles race for CAR T-cell therapy in digestive tract cancer. Immunol Rev. (2023) 320:100–19. doi: 10.1111/imr.13273 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 171. Garcia-Robledo JE, Cabrera-Salcedo S, Brandauer AM, Romano F, Rengifo-Martinez J, Toro-Pedroza A, et al. Engineering the next generation of CAR T- cells: precision modifications, logic gates and universal strategies to overcome exhaustion and tumor resistance. Front Oncol. (2025) 15:1698442. doi: 10.3389/fonc.2025.1698442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 172. Zhang Q, Chiang Z-C, Weng H, Chen X, Zhao N, Yan C, et al. Enhancement of B7-H3 chimeric antigen receptor-T cell efficacy via the coexistence effect of IL-7, IL-15 and CCL19 for pancreatic and lung xenograft tumors. (2026) 245:117673. doi: 10.1016/j.bcp.2026.117673 [DOI] [PubMed] [Google Scholar]
- 173. Guirguis DA, Hasan F, Morris N, Alabd A, Tawfik PMS, Pramanik K, et al. Tumor tough, therapy smarter: Rethinking CAR-T for pancreatic cancer. Semin Oncol. (2025) 52:152411. doi: 10.1016/j.seminoncol.2025.152411 [DOI] [PubMed] [Google Scholar]
- 174. Bains RS, Raju TG, Semaan LC, Block A, Yamaguchi Y, Priceman SJ, et al. Vascularized tumor-on-a-chip to investigate immunosuppression of CAR-T cells. Lab Chip. (2025) 25:2390–400. doi: 10.1039/d4lc01089b [DOI] [PubMed] [Google Scholar]
- 175. Krug A, Martinez-Turtos A, Verhoeyen E. Importance of T, NK, CAR T and CAR NK cell metabolic fitness for effective anti-cancer therapy: A continuous learning process allowing the optimization of T, NK and CAR-based anti-cancer therapies. Cancers. (2021) 14(1):183. doi: 10.3390/cancers14010183 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 176. Hou Y, Hu S, Liu C, Chen X, Wang Y, Li Y, et al. Beyond CAR-T Cells: exploring CAR-NK, CAR-M, and CAR-γδ T strategies in solid tumor immunotherapy. Front Immunol. (2025) 16:1675807. doi: 10.3389/fimmu.2025.1675807 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 177. Feng Y, Yang Z, Zhou Y, Liang Y, Zhao H. From innate-like to innate: the next wave of off-the-shelf CAR immunotherapies. Front Immunol. (2025) 16:1691743. doi: 10.3389/fimmu.2025.1691743 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 178. Redondo-Frutos RA, Justicia-Lirio P, Barbero-Jimenez C, Garnica M, Trudu L, Rodriguez-Marquez P, et al. Genetic engineering in CAR T cells for solid tumors: current state, barriers and future developments. Hum Gene Ther. (2025) 36:1138–53. doi: 10.1177/10430342251372041 [DOI] [PubMed] [Google Scholar]
- 179. Birocchi F, Almazan AJ, Parker A, Bouffard AA, Goncalves S, Kelly C, et al. On-target off-tumor toxicity of claudin18.2-directed CAR-T cells in preclinical models. Nat Commun. (2025) 16:9650. doi: 10.1038/s41467-025-61858-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 180. An M, Yan J, Liu B, Liu Q. Harnessing the power of CAR-NK cells for solid tumors: challenges, innovations, and future frontiers in immunotherapy. Cancer Commun (London England). (2026) 46:0023. doi: 10.34133/cancomm.0023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 181. Al Shaer NSH, Masoud I, Tleyjeh A, Al-Attas AA, Al-Attas MA, Khan MI, et al. CAR-T cells in solid tumors: engineering, biomarkers, translational pathways and the road ahead. Front Immunol. (2026) 17:1796675. doi: 10.3389/fimmu.2026.1796675 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 182. Ma L, Hostetler A, Morgan DM, Maiorino L, Sulkaj I, Whittaker CA, et al. Vaccine-boosted CAR T crosstalk with host immunity to reject tumors with antigen heterogeneity. Cell. (2023) 186:3148–3165.e20. doi: 10.1136/jitc-2023-sitc2023.0286 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 183. Hagel KR, Arafeh R, Gang S, Arnoff TE, Larson RC, Doench JG, et al. Systematic interrogation of tumor cell resistance to chimeric antigen receptor T-cell therapy in pancreatic cancer. Cancer Res. (2023) 83:613–25. doi: 10.1158/0008-5472.CAN-22-2245 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 184. Caruana I, Savoldo B, Hoyos V, Weber G, Liu H, Kim ES, et al. Heparanase promotes tumor infiltration and antitumor activity of CAR-redirected T lymphocytes. Nat Med. (2015) 21:524–9. doi: 10.1038/nm.3833 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 185. Lo A, Wang LS, Scholler J, Monslow J, Avery D, Newick K, et al. Tumor-promoting desmoplasia is disrupted by depleting FAP-expressing stromal cells. Cancer Res. (2015) 75:2800–10. doi: 10.1158/0008-5472.can-14-3041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 186. Das S, Valton J, Duchateau P, Poirot LJ. Stromal depletion by TALEN-edited universal hypoimmunogenic FAP-CAR T cells enables infiltration and anti-tumor cytotoxicity of tumor antigen-targeted CAR-T immunotherapy. (2023) 14:1172681. doi: 10.3389/fimmu.2023.1172681 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 187. Zhong W, Xiao Z, Qin Z, Yang J, Wen Y, Yu Z, et al. Tumor-derived small extracellular vesicles inhibit the efficacy of CAR T cells against solid tumors. Cancer Res. (2023) 83:2790–806. doi: 10.1158/0008-5472.can-22-2220 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 188. Jung IY, Narayan V, McDonald S, Rech AJ, Bartoszek R, Hong G, et al. BLIMP1 and NR4A3 transcription factors reciprocally regulate antitumor CAR T cell stemness and exhaustion. Sci Transl Med. (2022) 14:eabn7336. doi: 10.1126/scitranslmed.abn7336 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 189. Hammer Q, Perica K, Mbofung RM, van Ooijen H, Martin KE, Momayyezi P, et al. Genetic ablation of adhesion ligands mitigates rejection of allogeneic cellular immunotherapies. Cell Stem Cell. (2024) 31:1376–1386.e8. doi: 10.1016/j.stem.2024.06.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 190. Liu F, Tarannum M, Zhao Y, Zhang YJ, Ham JD, Lei K, et al. Selective HLA knockdown and PD-L1 expression prevent allogeneic CAR-NK cell rejection and enhance safety and anti-tumor responses in xenograft mice. Nat Commun. (2025) 16:8809. doi: 10.1038/s41467-025-63863-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 191. Jiang Q, Yu W, Ma J, Zhao M, Zou J, Mir S, et al. Robust differentiation of NK cells from MSLN.CAR-IL-15-engineered human iPSCs with enhanced antitumor efficacy against solid tumors. Sci Adv. (2025) 11:eadt9932. doi: 10.1126/sciadv.adt9932 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 192. Hecht JR, Molina JR, Liechty K, Welling TH, Grierson PM, Patel SP, et al. BASECAMP-1 screening study: a model for efficient enrolment in precision oncology clinical trials. BMJ Oncol. (2026) 5:e001033. doi: 10.1136/bmjonc-2025-001033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 193. Manry D, Bolanos K, DiAndreth B, Mock JY, Kamb A. Robust in vitro pharmacology of Tmod, a synthetic dual-signal integrator for cancer cell therapy. Front Immunol. (2022) 13:826747. doi: 10.3389/fimmu.2022.826747 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 194. Ouladan S, Orouji E. Chimeric antigen receptor-T cells in colorectal cancer: pioneering new avenues in solid tumor immunotherapy. J Clin Oncol Off J Am Soc Clin Oncol. (2025) 43:994–1005. doi: 10.1200/jco-24-02081 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 195. Tousley AM, Rotiroti MC, Labanieh L, Rysavy LW, Kim WJ, Lareau C, et al. Co-opting signalling molecules enables logic-gated control of CAR T cells. Nature. (2023) 615:507–16. doi: 10.1038/s41586-023-05778-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 196. Hyrenius-Wittsten A, Su Y, Park M, Garcia JM, Alavi J, Perry N, et al. SynNotch CAR circuits enhance solid tumor recognition and promote persistent antitumor activity in mouse models. Sci Transl Med. (2021) 13(591):eabd8836. doi: 10.1126/scitranslmed.abd8836 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 197. Zhu X, Chen J, Li W, Xu Y, Shan J, Hong J, et al. Hypoxia-responsive CAR-T cells exhibit reduced exhaustion and enhanced efficacy in solid tumors. Cancer Res. (2024) 84:84–100. doi: 10.1158/0008-5472.CAN-23-1038 [DOI] [PubMed] [Google Scholar]
- 198. Li D, Guo X, Yang K, Yang Y, Zhou W, Huang Y, et al. EpCAM-targeting CAR-T cell immunotherapy is safe and efficacious for epithelial tumors. (2023) 9:. doi: 10.1126/sciadv.adg9721 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 199. Chen J, López-Moyado IF, Seo H, Lio CJ, Hempleman LJ, Sekiya T, et al. NR4A transcription factors limit CAR T cell function in solid tumours. Nature. (2019) 567:530–4. doi: 10.1038/s41586-019-0985-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 200. Ghobadi A, Bachanova V, Patel K, Park JH, Flinn I, Riedell PA, et al. Induced pluripotent stem-cell-derived CD19-directed chimeric antigen receptor natural killer cells in B-cell lymphoma: a phase 1, first-in-human trial. Lancet (London England). (2025) 405:127–36. doi: 10.1016/s0140-6736(24)02462-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 201. Giordano Attianese GMP, Ash S, Irving M. Coengineering specificity, safety, and function into T cells for cancer immunotherapy. Immunol Rev. (2023) 320:166–98. doi: 10.1111/imr.13252 [DOI] [PubMed] [Google Scholar]
- 202. Yang DD, Macmorland W, Arnold JN. Current strategies for armoring chimeric antigen receptor T-cells to overcome barriers of the solid tumor microenvironment. Front Immunol. (2025) 16:1643941. doi: 10.3389/fimmu.2025.1643941 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 203. Biederstädt A, Rezvani K. Engineered natural killer cells for cancer therapy. Cancer Cell. (2025) 43:1987–2013. doi: 10.1016/j.ccell.2025.09.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 204. Qin D, Lei Y, Shu P, Zhang Y, Loh YH, Wang Y, et al. Supercharging CAR-T cells through transcriptional and epigenetic armoring. Theranostics. (2025) 15:3345–67. doi: 10.7150/thno.107908 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 205. Wang Y, Tong C, Dai H, Wu Z, Han X, Guo Y, et al. Low-dose decitabine priming endows CAR T cells with enhanced and persistent antitumour potential via epigenetic reprogramming. Nat Commun. (2021) 12:409. doi: 10.1038/s41467-020-20696-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 206. Hu J. Recent advances in chemokine-directed strategies to improve CAR-T cell infiltration into solid tumors. Int Immunopharmacol. (2026) 181:116783. doi: 10.1016/j.intimp.2026.116783 [DOI] [PubMed] [Google Scholar]
- 207. Zhang Y, Sun H, Zhao L, Zhao N, Chen Z, He L, et al. Barriers and strategies to enhance CAR-T cell infiltration in solid tumours: A systematic review. Immunology. (2026) 178:62–78. doi: 10.1111/imm.70094 [DOI] [PubMed] [Google Scholar]
- 208. Manshadi SM, Mehravar M, Hajifathali A, Roshandel E, Kaviani Jabali S. Induced pluripotent stem cell-derived NK cells as an off-the-shelf platform for cancer immunotherapy: opportunities, challenges, and clinical translation. Stem Cell Rev Rep. (2026) 22(6):2671–90. doi: 10.1007/s12015-026-11156-y [DOI] [PubMed] [Google Scholar]
- 209. Wen P, Ai Q, Fan X, Gu Z, Chen B, Qian H. Programmable smart CAR-T design: a new paradigm in precision immunotherapy driven by logic gates, conditional activation, allogeneic strategies, and artificial intelligence. Cancer Lett. (2026) 640:218257. doi: 10.1016/j.canlet.2026.218257 [DOI] [PubMed] [Google Scholar]
- 210. Bot A, Scharenberg A, Friedman K, Guey L, Hofmeister R, Andorko JI, et al. In vivo chimeric antigen receptor (CAR)-T cell therapy. Nat Rev Drug Discov. (2026) 25:116–37. doi: 10.1038/s41573-025-01291-5 [DOI] [PubMed] [Google Scholar]
- 211. Yin Q, Liang X, Zhang C, Bai Y, Li J, Qian F, et al. Lipid nanoparticle-mediated in vivo generation of panCAR immune cells for solid tumor immunotherapy. PNAS. (2026) 123:e2509698123. doi: 10.1073/pnas.2509698123 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 212. Liu Y, Sun Y, Wang P, Li S, Dong Y, Zhou M, et al. FAP-targeted CAR-T suppresses MDSCs recruitment to improve the antitumor efficacy of claudin18.2-targeted CAR-T against pancreatic cancer. J Transl Med. (2023) 21:255. doi: 10.1186/s12967-023-04080-z [DOI] [PMC free article] [PubMed] [Google Scholar]
