Organoid has emerged as the novel platform for preclinical anticancer drug testing in pancreatic ductal adenocarcinoma (PDA). However, most organoid models are not reconstituted with a tumor microenvironment. This study aimed to simultaneously establish PDA organoids and cancer-associated fibroblasts (CAFs) using endoscopic ultrasound-guided fine needle biopsy (EUS-FNB) samples.

Despite recent advances in molecular diagnostics and systemic cancer treatment, pancreatic ductal adenocarcinoma (PDA) still shows a late-stage presentation and has a lethal prognosis, with a 5-year survival of 9% in the United States and 12.2 % in South Korea. PDA is featured with abundant desmoplastic tumor stroma derived primarily from cancer-associated fibroblast (CAF), the most effective cell within the tumor microenvironment (TME). CAFs modulate cancer invasion and metastasis through extracellular matrix remodeling, holistic signaling interplay with cancer cells by soluble secreted factors, and crosstalk with infiltrating immune cells. Therefore, CAFs are an indispensable factor in understanding PDA biology.

In cancer precision medicine, organoid technology which is three-dimensional culture models grown from human cancer stem cells has recently emerged as a promising drug screening platform for standard and novel therapeutics, because it recapitulates biological features and genomic heterogeneity of original cancer. However, most current organoid models are not reconstituted with an intact TME, and the lack of a TME risks biasing tumor biology, leading to a phenotypic discrepancy between the organoid model and the original tumor.

To overcome this limitation, the co-culture of cancer organoids with various TME elements, the so-called mixed organoid, is being investigated. Representatively, the patient-derived organoid models comprising tumor epithelium and endogenous tumor-infiltrating immune cells including T, B, NK cells, and macrophages were established in diverse cancer by means of an air-liquid interface culture system. This model is promising as a preclinical screening platform for novel immune therapies including immune checkpoint inhibitors. In the bladder cancer field, a new concept-cancer organoid, named bladder cancer 'assembloids', comes into the spotlight. In this organoid platform, bladder tumor organoids were three-dimensionally reconstituted with multiple stromal components including CAFs, endothelial cells, immune cells, and outer muscle layer, forming a mature bladder-like layered structure. This model demonstrated that the tumor stroma represented by CAFs prevents the shift of the tumor subtype of the organoid models to a subtype different from that of the original tumor. Recently, the investigator of this study developed a new pancreatic cancer organoid model which is integrated with fibrous TME using CAFs. This CAF-integrated pancreatic cancer organoid model retained similar genetic and pathological characteristics to those in matched human cancer tissue. In this model, it was demonstrated that CAF-cancer cell interaction promotes epithelial-mesenchymal transition of cancer cells which is known to enhance cancer metastasis. Moreover, CAFs-induced extracellular matrix deposition impairs drug delivery to cancer cells. Hence, co-culturing cancer cells, as well as CAFs, is an imperatively necessary strategy to establish a reliable preclinical organoid model for cancer precision medicine.

To generate PDA mixed organoid model, the acquisition of sufficient cancer and stromal tissue is a prerequisite but a demanding process. This is because 80% of the PDA patients are not operative candidates attributed to locally advanced tumor burden or systemic spread. Therefore, the resource of PDA organoid models is largely derived from endoscopic ultrasound-guided fine needle aspiration/biopsy (EUS-FNA/B). The creation of a mixed PDA organoid model from EUS-guided biopsy samples would be challenging because the amount of total tissue materials is smaller than the surgical samples, and EUS-derived samples usually contain relatively limited stromal tissue compared with cancer cells.

Hence, this study is aimed to simultaneously establish the patient-derived PDA organoids as well as CAFs using EUS-FNB samples.

Inclusion Criteria:

• solid pancreatic mass lesion presumed PDA on cross-sectional imaging, regardless of tumor resectability

Exclusion Criteria:

• inability to provide informed consent, or coagulopathy (international normalized ratio >1.5, platelet count <50,000 per mcL)