Breakthroughs in CAR-T Cell Therapy for Colorectal Cancer

Last month, March 2025, marked National Colorectal Cancer Awareness Month, a time to reflect on the progress made in the fight against colorectal cancer (CRC) and to spotlight emerging therapeutic innovations. As the third most commonly diagnosed cancer worldwide and the second leading cause of cancer-related deaths, CRC continues to pose a significant public health burden. While traditional treatment strategies such as surgery, chemotherapy, and radiotherapy remain mainstays, metastatic colorectal cancer (mCRC) often proves resistant to these modalities, highlighting an urgent need for novel therapeutic approaches.

Among these, adoptive cell therapy-particularly chimeric antigen receptor T-cell (CAR-T) therapy-has shown remarkable success in hematologic malignancies and is now being explored for solid tumors, including CRC. This review explores recent advancements in CAR-T cell therapy for CRC, highlighting emerging therapeutic targets, current clinical trial outcomes, and innovative genetic engineering strategies designed to enhance efficacy and address the specific challenges associated with solid tumors like CRC.

Colorectal Cancer: The Need for Novel Therapies

Colorectal cancer affects approximately 1 in 23 men and 1 in 25 women globally, with an estimated 1.8 million new cases and 881,000 deaths annually (1). While early-stage CRC is often curable through surgical resection, outcomes for mCRC remain dismal—stage IV disease carries a five-year survival rate of only 14%, compared to over 90% in stage I (2).

This sharp decline in survival rates highlights the limitations of conventional therapies and underscores the biological complexities of treating solid tumors such as CRC. One of the most formidable obstacles is the tumor microenvironment (TME), which in CRC is notably immunosuppressive. It is enriched with regulatory T cells, myeloid-derived suppressor cells (MDSCs), and immunosuppressive cytokines like TGF-β and IL-10, all of which dampen effective antitumor immune responses. Moreover, CRC tumors exhibit significant antigenic heterogeneity, often driven by common mutations in oncogenes and tumor suppressor genes such as KRAS, BRAF, and TP53. This heterogeneity allows subclonal populations to escape immune surveillance. Compounding these issues, the dense stromal architecture and poor vascularization of CRC tumors physically restrict the infiltration of both therapeutics and immune effector cells, further diminishing the efficacy of existing treatments (3,4).

Immune checkpoint inhibitors (ICIs) have demonstrated significant efficacy in a molecularly distinct subset of colorectal cancers characterized by mismatch repair deficiency (dMMR) or high microsatellite instability (MSI-H). These tumors, which account for approximately 15–18% of stage II and only 4–5% of stage IV CRC cases, are marked by a high tumor mutational burden and abundant neoantigen expression, rendering them highly immunogenic. As a result, they respond well to ICIs such as pembrolizumab, which has been approved as a first-line treatment in this population (5). Despite this success, the overwhelming majority of metastatic CRCs (around 95%) are microsatellite stable (MSS) and proficient in mismatch repair (pMMR). These tumors are considered immunologically “cold,” with low levels of T cell infiltration and minimal neoantigen presentation. Consequently, they exhibit poor responsiveness to ICIs, leaving patients with limited treatment options beyond conventional chemotherapy and targeted agents. (6-8). In addition, mutations in the KRAS gene—which occur in approximately 40–50% of CRC patients—render tumors resistant to therapies, eliminating a key line of treatment for a large patient subgroup (1). This resistance, alongside limited responses to immune checkpoint inhibitors in pMMR/MSS tumors, combined with the systemic toxicity and transient efficacy of conventional chemotherapy, further highlights the urgent need for innovative and targeted strategies in CRC (9).

CAR-T Therapy: A Precision Approach for CRC

CAR-T cells, derived from a patient’s peripheral blood, are genetically engineered to express chimeric antigen receptors that target tumor-associated antigens. These synthetic receptors incorporate both antigen-recognition domains and intracellular costimulatory motifs (e.g., CD28, 4-1BB) to boost T cell activation, persistence, and cytotoxic function. This approach enables the redirected T cells to selectively recognize and eliminate cancer cells, offering a personalized and potent alternative to conventional cancer therapies. Lentiviral vectors (LVVs)  are commonly employed to stably introduce CAR constructs into T cells. In the context of solid tumors—such as CRC—further genetic enhancements, including inducible cytokine expression (e.g., IL-12, IL-15), resistance to immunosuppressive signals, or built-in checkpoint inhibition, are actively being explored to overcome hostile tumor microenvironments and improve therapeutic outcomes.

Unlike immune checkpoint inhibitors (ICIs), which largely benefit only a small subset of dMMR/MSI-H CRCs, CAR-T cell therapy holds the potential to overcome immune exclusion and be effective even in immunologically “cold” tumors such as microsatellite stable (MSS) CRC (10). With growing interest in overcoming the immunosuppressive tumor microenvironment and improving antigen targeting strategies, ongoing translational research and early-phase clinical trials are rapidly advancing CAR-T platforms. These efforts aim to tailor adoptive cell therapies for CRC patients with limited responses to standard treatments and historically poor prognoses.

Designing CAR-T for CRC: Evolving Generations and Emerging Targets

CAR-T cell therapy has advanced significantly over the past two decades, progressing through four distinct generations to improve efficacy, persistence, and adaptability in challenging tumor environments:

• First-generation CARs featured only the CD3ζ signaling domain without additional co-stimulatory signals, leading to suboptimal T cell activation, limited in vivo persistence, and modest anti-tumor responses.
• Second-generation CARs, now the backbone of most FDA-approved therapies (e.g., anti-CD19 CAR-T products), incorporate one co-stimulatory domain such as CD28 or 4-1BB. These additions greatly enhance T cell proliferation, persistence, and cytotoxic function.
• Third-generation CARs combine two co-stimulatory domains (commonly CD28 and 4-1BB or OX40), aiming for synergistic activation. While this design offers more potent anti-tumor activity, it may also increase the risk of cytokine release syndrome (CRS) and other immune-related toxicities.
• Fourth-generation CARs, also known as TRUCKs (T cells Redirected for Universal Cytokine-mediated Killing), integrate inducible expression of pro-inflammatory cytokines like IL-12, IL-15, or IL-18. These modifications aim to overcome the immunosuppressive TME, recruit innate immune cells, and enhance local immune responses within solid tumors like CRC (Figure 1).

Figure 1. CAR Structure across four generations.

A major challenge in developing CAR-T therapies for CRC lies in identifying suitable tumor-associated antigens (TAAs) that are highly expressed on cancer cells but minimally present on healthy tissues to minimize off-target toxicity. Several promising TAAs have emerged from preclinical and early clinical studies (Figure 2) (11):

• CD166 (ALCAM): CD166 is a transmembrane glycoprotein involved in cell adhesion and immune regulation. It is overexpressed in CRC and implicated in tumor progression. CAR-T strategies are exploring the interaction between CD166 and the SCRC3 domain of CD6, which binds selectively to CD166. This interaction may enhance antigen-specific targeting and T cell activation, making CD166 a viable candidate for CAR engagement in CRC.
• Nectin4: Nectin4 is a member of the nectin family of immunoglobulin-like adhesion molecules and is highly expressed in various solid tumors, including CRC. It has limited expression in normal adult tissues, making it an attractive TAA. Innovative strategies are evaluating dual targeting of Nectin4-positive tumor cells and FAP-positive cancer-associated fibroblasts (CAFs), using a combination of Nectin4-7.19 CAR-T and FAP-12 CAR-T cells. This dual approach aims to disrupt both tumor cells and the stromal barrier, potentially enhancing treatment efficacy (12).
• CEA (Carcinoembryonic Antigen): CEA is one of the most well-established biomarkers in CRC, widely used for diagnostic and monitoring purposes. It is overexpressed in most CRC cases and has become a leading target for CAR-T development. Early-phase clinical trials have demonstrated that CEA-targeted CAR-T therapies are generally well-tolerated and capable of inducing tumor regression in some patients, even at high doses. Strategies to mitigate on-target off-tumor effects in normal tissues are ongoing.
• Guanylyl Cyclase C (GUCY2C): GUCY2C is a membrane-bound receptor involved in regulating intestinal epithelial cell homeostasis. It is selectively expressed on the apical surface of intestinal epithelial cells and overexpressed in primary and metastatic CRC, but is largely inaccessible to systemic circulation in normal tissues. Preclinical models using GUCY2C-targeted CAR-T cells have demonstrated potent and selective antitumor activity, with minimal toxicity, making it a compelling candidate for further development.
• CD318 (CDCP1): CD318 is a transmembrane glycoprotein associated with increased metastatic potential and poor prognosis in colorectal and other epithelial cancers. Early studies suggest that CAR-T cells targeting CD318 can effectively recognize and eliminate CRC cells, supporting its potential as a therapeutic target. Further investigations are needed to assess its expression heterogeneity and safety profile in clinical settings.
• Globotriaosylceramide (Gb3/CD77): Gb3 is a glycolipid overexpressed in several solid tumors, including CRC, and is linked to tumor invasiveness, angiogenesis, and metastasis. It is particularly enriched in poorly differentiated tumors. Lectin-based CAR constructs targeting Gb3 have shown antigen-specific cytotoxicity against CRC cell lines in preclinical studies. Ongoing research is evaluating the safety and selectivity of these lectin-based CARs to avoid potential toxicity to normal Gb3-expressing tissues.

Figure 2. CAR-T targets are being investigated.

PackGene supports CAR-T development by providing LVV production services of CARs targeting these and other antigens (EpCAM, CDH17, etc). Continued innovation in CAR construct design and antigen selection will be vital to enhance efficacy while minimizing off-tumor effects.

Key CAR-T Clinical Trials in CRC

Although CAR-T therapy in colorectal cancer (CRC) remains in earlier stages compared to its success in hematologic malignancies, several clinical trials have reported encouraging signals of safety and efficacy (Table 1):

•  CEA-targeted CAR-T Cells (Phase I):

In a Phase I trial conducted in China, 10 patients with metastatic CRC received repeated infusions of autologous CAR-T cells targeting CEA. The treatment was generally well-tolerated. Four patients achieved stable disease, while two patients exhibited significant declines in serum CEA levels and tumor burden on imaging. Importantly, this trial did not report high-grade CRS or neurotoxicity, suggesting that regional administration (such as hepatic artery infusion) may mitigate systemic toxicity while maintaining localized antitumor activity (13).

• GUCY2C CAR-T Cells (IM96, ImmunoChina):

In the IM96 trial, CAR-T cells targeting GUCY2C demonstrated an overall response rate (ORR) of 26.3% across all dose levels. Notably, the high-dose cohort (DL3) showed a 40% ORR and a median progression-free survival (PFS) of 7 months. These findings highlight the potential for GUCY2C to serve as a tumor-specific and relatively safe target in CRC CAR-T therapy.

•  GCC19CART (NCT05319314):

Conducted in the U.S., this trial evaluated GUCY2C-targeted CAR-T therapy in patients with advanced gastrointestinal malignancies, including CRC. At dose level 1, the study reported a 50% ORR with manageable safety profiles. Cytokine release syndrome was mild (Grade 1–2), and no neurotoxicity or off-tumor effects were observed. This study reinforced the viability of targeting GUCY2C in Western populations, supporting further dose escalation and combination strategies.

•  A2B694 CAR-T (EVEREST-2, NCT06051695):

This novel trial by A2 Biotherapeutics uses a logic-gated CAR-T system, specifically engineered to enhance tumor specificity. The CAR targets mesothelin, a glycoprotein overexpressed in CRC and other solid tumors, while simultaneously using a Tmod inhibitory receptor to recognize cells with loss of heterozygosity (LOH) at HLA-A*02, a tumor-specific genetic feature. This dual-signal mechanism allows CAR-T cells to remain inert in normal tissues while becoming fully activated in tumor cells. Though still in early phases, the approach represents a leap toward improving safety and minimizing on-target off-tumor effects.

•  P-MUC1C-ALLO1 (Poseida Therapeutics):

Unlike traditional autologous CAR-T therapies, this study explores an allogeneic off-the-shelf product targeting MUC1-C, a cleavage-resistant form of MUC1 that is selectively expressed in cancer cells. The product incorporates gene-editing technologies to eliminate endogenous T cell receptors, reducing the risk of Graft-versus-Host Disease (GvHD). Initial trial data indicated no dose-limiting toxicities, no graft-versus-host disease, and promising safety even at higher doses. The trial is ongoing, but the use of allogeneic products could dramatically enhance the scalability and accessibility of CAR-T therapies for CRC.

Collectively, these early-phase trials highlight the growing interest in CAR-T therapy for colorectal cancer and underscore its clinical feasibility, manageable safety profile, and encouraging signs of antitumor efficacy, paving the way for further development and optimization in this challenging solid tumor setting.

Table 1. Key CRC CAR-T trials.

Challenges and Future Directions

Despite encouraging advances in the application of cell therapies for CRC, several critical challenges must be addressed to fully realize their therapeutic potential:

• Identifying Optimal Tumor-Specific Antigens: The lack of ideal targets raises concerns about on-target, off-tumor toxicity and limits the applicability of current CAR designs. More novel targets need to be explored.
• Overcoming the TME: Strategies that enhance T cell trafficking, persistence, and function in this hostile environment are urgently needed.
• Managing On-Target, Off-Tumor Toxicities: Because many CRC-associated antigens are also expressed at low levels on normal tissues, minimizing collateral tissue damage requires advanced CAR engineering and stringent preclinical validation.
• Scalable and Cost-Effective Manufacturing: The complexity and cost of producing autologous cell therapies remain a major barrier to widespread clinical implementation. Innovations in allogeneic platforms and closed-system manufacturing are actively being explored.
• Ensuring Long-Term Efficacy and Safety: Durable clinical responses and long-term safety profiles remain to be established. Extended follow-up and real-world studies are essential to monitor delayed toxicities and relapse rates

To overcome these barriers, current research is focusing on the key strategies to improve CAR-T therapy’s durability, safety, and efficacy. Researchers are increasingly leveraging next-generation genetic engineering strategies (14):

• Checkpoint Inhibition: CAR-T cells can be genetically modified using tools like CRISPR-Cas9 to knock out immune checkpoint molecules such as PD-1, CTLA-4, and LAG-3, which are associated with T cell exhaustion in the TME. Emerging targets such as FOXP3, a key regulator of regulatory T cell function, and SOCS family proteins, which negatively regulate cytokine signaling, are also under investigation to further enhance CAR-T persistence and antitumor function.
• Armored CAR-T Cells: Also known as TRUCKs (T cells Redirected for Universal Cytokine Killing), these are engineered to secrete immunostimulatory cytokines such as IL-15 or IL-18, which can promote T cell persistence, stimulate local immune activation, and modulate the suppressive TME. Additional enhancements, such as co-expression of chemokine receptors (e.g., CXCR2) enable improved tumor homing and infiltration.
• Metabolic Reprogramming: The harsh metabolic environment of solid tumors limits CAR-T function. Engineering approaches that improve mitochondrial biogenesis and oxidative metabolism-for instance, by upregulating FOXO1—can improve CAR-T cell longevity. Similarly, knocking out diacylglycerol kinase isoforms enhances resistance to metabolic inhibition and supports sustained T cell activation in nutrient-poor environments.

Together, these cutting-edge modifications aim to overcome the multifaceted challenges of solid tumors like CRC and may ultimately help transition CAR-T therapy from a niche modality to a broader platform applicable across various tumor types.

Conclusion

As National Colorectal Cancer Awareness Month March 2025 brings renewed focus to this devastating disease, it also highlights the rapid evolution of therapeutic strategies. While conventional treatments remain essential, they are often insufficient for advanced-stage CRC. CAR-T therapy, with its evolving designs and expanding clinical trials, represents a paradigm shift—offering personalized, genetically engineered solutions capable of overcoming the immune resistance and heterogeneity of solid tumors. Although still in early stages for CRC, the growing body of preclinical and clinical evidence suggests that CAR-T therapies-particularly when combined with advanced engineering strategies-hold great promise for improving outcomes. Continued collaboration between researchers, clinicians, and biotech innovators will be critical to transforming these therapies from experimental to essential.

References

1. Bray, F., et al. (2024). CA: A Cancer Journal for Clinicians.
   https://acsjournals.onlinelibrary.wiley.com/doi/10.3322/caac.21834;
2. Ilyas, M.I.M., et al. (2023). Clinics in Colon and Rectal Surgery.
   https://doi.org/10.1055/s-0043-1761447
3. Waldner, M.J., et al. (2023). Cells.
   https://doi.org/10.3390/cells12081139
4. Brás, M.M., et al. (2022). Cancers.
   https://doi.org/10.3390/cancers14081945
5. André, T., et al. (2020). New England Journal of Medicine.
   https://doi.org/10.1056/NEJMoa2017699
6. Yan, S., et al. (2024). Frontiers in Immunology.
   https://doi.org/10.3389/fimmu.2024.1403533
7. Zocche, D.M., et al. (2015). Frontiers in Genetics.
   https://doi.org/10.3389/fgene.2015.00116
8. Lorenzi, M., et al. (2020). Journal of Oncology.
   https://doi.org/10.1155/2020/1807929
9. Dosunmu, G.T., et al. (2024). Genes.
   https://doi.org/10.3390/genes15050538
10. Liu, Y.T., et al. (2021). Theranostics.
    https://doi.org/10.7150/thno.58390
11. Kamrani, A., et al. (2024). Cell Communication and Signaling.
    https://doi.org/10.1186/s12964-023-01430-8
12. Li F, et al. (2022). Frontiers in Immunology.
    https://doi.org/10.3389/fimmu.2022.958082
13. Ouladan, S., et al. (2025). Journal of Clinical Oncology.
    https://doi.org/10.1200/JCO-24-02081
14. Albarrán-Fernández, V., et al. (2025). Nature Communications.
    https://doi.org/10.1038/s41467-025-56527-0