This example study evaluates the safety, tolerability, and preliminary anti-tumor activity of investigational, dual-targeting chimeric antigen receptor natural killer (CAR-NK) cell products for patients with advanced pancreatic ductal adenocarcinoma (PDAC). Participants are assigned to one of two biomarker-defined cohorts based on tumor antigen expression: (A) Mesothelin (MSLN) and/or MUC1, or (B) Claudin 18.2 (CLDN18.2) and/or MUC1. The study uses a dose-escalation followed by dose-expansion design to define a recommended Phase 2 dose (RP2D) and to estimate response rates in each cohort.

• Rationale: PDAC is characterized by aggressive biology, antigen heterogeneity, and an immunosuppressive tumor microenvironment. Dual-targeting CAR designs aim to reduce antigen-escape by enabling recognition of either target antigen on tumor cells.
• Investigational products: Two off-the-shelf (allogeneic) CAR-NK products are evaluated. EB-DNK101 targets MSLN and MUC1. EB-DNK102 targets CLDN18.2 and MUC1. Both products are engineered to enhance persistence (e.g., membrane-bound or secreted IL-15; example) and incorporate an inducible safety switch (e.g., iCasp9; example).
• Target assessment and cohort assignment: Tumor tissue (archival or fresh biopsy) is tested centrally by immunohistochemistry (IHC) for MSLN, MUC1, and CLDN18.2. Participants are assigned to Arm A (MSLN/MUC1) or Arm B (CLDN18.2/MUC1) based on predefined positivity thresholds. If more than one cohort is eligible, assignment prioritizes the strongest antigen expression and product availability.
• Conditioning and administration: Participants receive lymphodepleting chemotherapy (fludarabine + cyclophosphamide; example regimen) followed by intravenous infusion of the assigned CAR-NK product.

Repeat infusions (up to 3 total) may be permitted in the absence of prohibitive toxicity and with at least stable disease.

• Safety monitoring: Participants are monitored closely for cytokine release syndrome (CRS), immune effector cell-associated neurotoxicity syndrome (ICANS), infusion reactions, and other adverse events.

Dose-limiting toxicities (DLTs) are assessed during the first 28 days after first infusion.

• Efficacy and biomarker assessments: Tumor response is assessed by imaging (RECIST v1.1) at regular intervals (e.g., every 8 weeks). Exploratory endpoints include CAR-NK expansion/persistence, cytokine profiling, and association of antigen density with response.
• Target down-selection plan , After completion of Part 1 and an initial subset of Part 2 expansion participants, an internal scientific review compares antigen prevalence, manufacturability, safety, and preliminary activity across cohorts to prioritize the lead dual-target construct for subsequent confirmatory development.

Inclusion Criteria:

• Age 18 to 75 years at the time of consent.
• Histologically or cytologically confirmed pancreatic ductal adenocarcinoma (PDAC).
• Unresectable locally advanced or metastatic disease with progression after at least 1 prior standard systemic therapy regimen, or intolerance/ineligibility for standard therapy.
• At least 1 measurable lesion per RECIST v1.1.
• Tumor antigen expression by central IHC (archival or fresh biopsy): • Arm A eligibility: MSLN positive and/or MUC1 positive. • Arm B eligibility: CLDN18.2 positive and/or MUC1 positive. (Example threshold: IHC 2+ or 3+ staining in >=50% of tumor cells, or H-score above protocol-defined cutoff.)
• ECOG performance status 0-1.
• Adequate organ function (example): ANC >= 1.0 x 10^9/L; platelets >= 75 x 10^9/L; hemoglobin >= 8 g/dL; AST/ALT <= 3x ULN (<= 5x ULN with liver metastases); total bilirubin <= 1.5x ULN; creatinine clearance >= 50 mL/min.
• Life expectancy >= 12 weeks.
• Negative pregnancy test for individuals of childbearing potential; agreement to use effective contraception during study participation and for a protocol-defined follow-up period.
• Ability to understand and willingness to sign written informed consent.

Exclusion Criteria:

• Active or untreated CNS metastases or carcinomatous meningitis.
• Clinically significant uncontrolled infection (including uncontrolled bacterial, fungal, or viral infection).
• Known active hepatitis B or hepatitis C with detectable viral load; known uncontrolled HIV infection
• Prior allogeneic hematopoietic stem cell transplant or solid organ transplant.
• Prior gene-modified cellular therapy (e.g., CAR-T/CAR-NK) within 6 months or prior therapy targeting the same antigen(s)

Open-label design due to the nature of cellular infusion and required safety monitoring.