The Challenge of Antigen Heterogeneity in PDAC

Pancreatic ductal adenocarcinoma (PDAC) presents a formidable challenge for cell-based therapies due to its dense, desmoplastic stroma and significant intra-tumoral heterogeneity. This biological complexity often leads to incomplete responses or relapse following treatment with single-antigen targeted agents, as tumor cells that lack the target antigen can evade therapy and proliferate.

A dual-antigen targeting strategy for CAR-NK cells is designed to overcome this limitation. By engaging two separate tumor-associated antigens, the therapy can achieve broader recognition of the cancer cell population and substantially reduce the potential for antigen-loss escape variants to emerge.

Validating Complex Constructs in Clinically Relevant Models

The selection of an appropriate in vivo model is fundamental to generating a predictive pharmacology dataset. While CDX models are useful for initial screening, PDX models offer superior clinical correlation for evaluating advanced cell therapies.

• Preservation of TME: PDX models maintain the complex stromal and cellular architecture of the original patient tumor.

• Genomic Stability: These models better reflect the genomic diversity and stability of the primary tumor.

• Predictive Power: Efficacy data from well-characterized PDX models has shown a higher correlation with clinical outcomes.

This approach aligns with principles for achieving durable therapeutic outcomes, where targeting the correct progenitor or heterogeneous cell populations is key to stable phenotype correction (PMID: 14577924). Our >100,000 sq ft facility is fully equipped to manage large-scale, complex PDX studies to support these programs.

Characterizing the In Vivo Response and Immune Microenvironment

Beyond primary efficacy, a successful preclinical program must thoroughly characterize the biological activity of the CAR-NK cell therapy. This includes assessing cell trafficking, persistence, and potential for off-tumor engagement. Understanding how the therapy interacts with local suppressive microenvironments is also vital for predicting clinical performance. The mechanisms of localized immune tolerance can dictate the success or failure of a non-self therapeutic, making microenvironment analysis a key dataset (PMID: 19575456).

This comprehensive data package is foundational to a successful regulatory submission. Since 2019, programs we have supported have achieved a 100% IND success rate, a testament to rigorous preclinical validation (the Franklin Biolabs brand was launched in 2024).