Here, we present current protocols for isolating, expanding and characterizing patient-derived PDAC organoids. Our current success rate of establishing organoid culture is over 70%; therefore, these methods have not yet been perfected and are expected to improve and evolve over time. Important consideration should be given to sample size, as PDAC has a low neoplastic cellularity. Consequently, small specimens will contain few tumor cells, and will only generate a handful of organoids. Additionally, many patients receive chemotherapy and/or chemoradiation-based neoadjuvant treatment prior to surgical intervention15. If the treatment is effective for a particular patient, the tumor tissue may be devoid of viable cells. Acquisition of chemo-naive patient samples is preferred for initial optimization of these methods, but this is not always possible. Interestingly we have found that ischemia time following surgical removal of the tumor tissue is not a major criterion for successful organoid isolation, as long as the sample is processed within 24 h.
Pancreatic ductal adenocarcinoma is a disease characterized by a strong desmoplastic reaction and deposition of a dense stromal matrix. While organoids are an excellent tool for the rapid isolation and expansion of the epithelial compartment, the model does not recapitulate the complex stroma of PDAC. Other methodologies such as patient derived xenografts16 or air liquid interface culture17 allow for a stromal compartment, however they may be challenging to expand quickly. When choosing a model system, the researcher should carefully consider the strengths and weaknesses of each6.
The heterogeneous biology of this disease impacts organoid establishment as some patient-derived organoids grow extremely well in our conditions while other are much slower by comparison. The protocols above describe a Wnt ligand rich condition to isolate and expand all patient-derived organoids, yet others have shown that some patient's tumors are able to grow in the absence of Wnt conditioned media11,12. Further testing will be required to determine if using a range of media conditions enhances the successful establishment of organoids, as was recently demonstrated for ovarian cancer organoids18. This multiplex approach is however limited by the low number of tumor cells that can be isolated from small patient samples. Additionally, normal untransformed ductal organoids can arise from an organoid isolation, particularly if the tumor tissue is adjacent to normal tissue9. To reduce the risk of normal organoid contamination, larger tissue samples can be subdivided into smaller independent fragments using morphological differences such as well vascularized (blood is visible) versus hypovascular regions, and hard nodules versus soft tissue.
The methods and protocols described here are the current standard approaches used in our laboratory for organoid isolation and they should be tested and adapted for each laboratory environment. For instance, the enzymatic dissociation (steps 2.6 to 2.9) of the tumor tissue is particularly important to optimize. Small equipment differences (nutator vs rotator mixer) can lead to significantly different timing for this step. Furthermore, the tissue dissociation can be fine-tuned by increasing or reducing the concentration of the Collagenase/Hyaluronidase mixture. Care must be taken to not treat all samples in the same manner. For example, in some cases organoids can be isolated from ascites fluid from advanced PDAC patients without mechanical or enzymatic dissociation.
DNA sequencing is the current gold standard to determine the presence or absence of tumor organoids as PDAC is driven by frequent mutations in KRAS, TP53, SMAD4 and CDKN2A. Transcriptomic analysis can reveal different tumor subtypes while pharmacotyping can uncover patient-specific therapeutic vulnerabilities9. These protocols enable PDAC researchers to develop their own library of patient-derived organoids and to profile the biology of these models.