CAR-T Cell Therapy for Pancreatic Cancer: Current Evidence, Biological Barriers, and Future Directions

Introduction: Why Pancreatic Cancer Is So Lethal

Pancreatic cancer is disproportionately lethal relative to how often it occurs. National Cancer Institute SEER data show it accounts for roughly 3.3% of new U.S. cancer cases but ranks third among causes of cancer death (behind lung and colorectal cancer). Overall five-year relative survival is about 13%, rising to roughly 44% for localized disease, 17% for regional disease, and only about 3% for distant disease. Median age at diagnosis is 71 years. Vague or absent early symptoms mean most patients are diagnosed after spread beyond the pancreas (National Cancer Institute, “Cancer Stat Facts”).

More than 90% of pancreatic cancers are exocrine, and most of those are PDAC. KRAS mutations occur in roughly 90% of PDAC tumors, with later inactivation of tumor suppressors such as TP53, CDKN2A, and SMAD4. That genetic profile helps explain resistance to many targeted therapies and motivates immunotherapy approaches that target surface proteins rather than correcting internal drivers (El Bejjani et al.). Incidence is higher in men than women and highest among Black patients; rates have been rising in several demographic groups, including younger women under 55 (Samaan et al.; National Cancer Institute, “SEER*Explorer”).

How CAR-T Therapy Works

CAR-T therapy collects a patient’s T cells by leukapheresis, genetically engineers them to express a chimeric antigen receptor, expands them in the laboratory, and reinfuses them—usually after lymphodepleting chemotherapy. The receptor recognizes a chosen surface antigen without relying on the major histocompatibility complex that tumors often evade. Binding activates the T cell to kill the target (El Bejjani et al.).

First-generation CARs used CD3-zeta alone and persisted poorly. Nearly all investigational products discussed for PDAC are second-generation designs that add one co-stimulatory domain (commonly CD28 or 4-1BB). Fourth-generation “armored” CARs that secrete cytokines or other payloads at the tumor site are of particular interest for PDAC’s hostile microenvironment. Manufacturing typically takes about two to eight weeks from leukapheresis to product delivery—a meaningful constraint for a rapidly progressive disease (“High Cost of Chimeric Antigen Receptor T-Cells”).

Biological Barriers in PDAC

Unlike B-cell malignancies where CAR-T has produced durable remissions, PDAC presents several compounding obstacles. An unusually dense desmoplastic stroma can physically block CAR-T cells from reaching tumor cells and may constitute the majority of tumor mass (El Bejjani et al.). The microenvironment recruits regulatory T cells and myeloid-derived suppressor cells, and poor perfusion creates hypoxia that impairs T-cell metabolism—pushing infiltrating cells toward exhaustion (“CAR-T Cell Therapy for Pancreatic Cancer”).

PDAC also lacks a single antigen expressed uniformly on all tumor cells, and many candidate antigens appear at lower levels on healthy tissue, raising on-target, off-tumor toxicity risk and limiting dosing. Even after infusion, cells must traffic to the tumor, infiltrate through stroma, and persist in an active state; early-phase trials have repeatedly identified weak persistence as a reason responses are partial or short-lived rather than durable.

Antigen Targets Under Investigation

Mesothelin is overexpressed in an estimated 80% of PDAC tumors while largely absent from most normal tissues beyond mesothelial lining. A related mesothelin-targeting T-cell receptor fusion construct (gavocabtagene autoleucel) reported, in a phase 1/2 trial of 32 patients with mesothelin-expressing solid tumors, an overall response rate of 20%, disease control rate of 77%, and six-month overall survival of 70%; grade 3-or-higher cytokine release syndrome occurred in 25% of patients (Adusumilli et al.).

CEA is elevated in an estimated 30–60% of PDAC cases. Because it is also present at low levels on healthy gastrointestinal epithelium, one trial delivered anti-CEA CAR-T cells into the hepatic artery rather than intravenously; six of eight enrolled patients completed the protocol (Katz et al., “Phase I”). A follow-on study with pressure-enabled delivery reported roughly 5.2-fold improvement in delivery to liver metastases versus standard infusion (Katz et al., “HITM-SURE”).

Claudin18.2 is largely restricted to stomach lining in healthy tissue but appears in pancreatic and other gastrointestinal cancers. In a phase 1 trial of CT041 in 37 patients with previously treated Claudin18.2-positive digestive cancers, overall response rate was 48.6% and disease control rate 73.0%; cytokine release syndrome was common but nearly all grade 1–2, with no grade 3-or-higher CRS, neurotoxicity, or treatment-related deaths reported. Those results led to a Phase 3 randomized trial, making Claudin18.2 one of the leading candidates for a gastrointestinal solid-tumor CAR-T approval (Qi et al.). HER2 and cell-surface GRP78 are also under study, and dual-targeting or combination designs aim to reduce escape by antigen-negative cells.

Clinical Trial Landscape

Most PDAC-directed CAR-T trials remain early-phase. Representative registered studies include mesothelin programs (e.g., NCT03907852 with ORR 20% and DCR 77% in mixed solid tumors; NCT06054308 neoadjuvant feasibility in resectable PDAC), hepatic-artery CEA trials (NCT02416466, NCT02850536), and Claudin18.2 trials including phase 1 NCT03874897 and ongoing phase 3 NCT07103668 (National Library of Medicine). Safety has generally been manageable, but efficacy outside Claudin18.2 programs remains limited. Emerging strategies include cytokine-secreting armored CARs, dominant-negative PD-1 receptors, dual-antigen logic, combinations with chemotherapy or checkpoint inhibitors, and point-of-care manufacturing to shorten timelines and costs (El Bejjani et al.; Palani et al.).

Cost, Access, and Equity

Approved blood-cancer CAR-T products typically list above $400,000 per patient; real-world total costs of care often center near $600,000 and can exceed $1 million, and roughly two-thirds of health plans restrict coverage for cellular and gene therapies (“High Cost of Chimeric Antigen Receptor T-Cells”). Access is uneven: geographic and racial disparities appear in CAR-T trial enrollment for other cancers, and rural or smaller systems often lack specialized infrastructure (Alqazaqi et al.). Because pancreatic cancer incidence is already highest among Black patients, a future PDAC CAR-T that follows existing access patterns risks inequitable benefit. Point-of-care manufacturing in one India demonstration reported costs near $35,000 versus over $400,000 for comparable U.S. commercial products (Palani et al.), but such models are not yet standard practice in the United States.

Conclusion

Pancreatic cancer’s high mortality relative to incidence—and the demographic disparities in that burden—make it a priority for new therapies. CAR-T is scientifically compelling but has so far been limited in PDAC by stroma, microenvironment, antigen heterogeneity, and persistence. Near-term promise lies more in armored, dual-antigen, and combination strategies than in any single-antigen product alone. Scientific success will still require parallel progress on manufacturing cost, reimbursement, and health-system infrastructure if benefit is to reach patients equitably.

Works Cited