Conventional 2-D cell culture models are essential tools in cancer research for studying cancer progression and therapy1. These models not only allow for controlled experimental conditions to investigate the molecular and cellular mechanisms underlying cancer development and progression but also provide cost-effective and relatively rapid experimental results. Their usage, however, is restricted due to the limited cellular diversity in the tumor microenvironment (TME), which cannot be recapitulated in a coplanar nature of mono-culture models2. Additionally, cell culture models offer an oversimplified environment compared to the human body3. Thus, to better understand cancer progression and develop effective treatments, researchers have turned to innovative models that aim to recapitulate the TME in nature closely. Consequently, maintaining fidelity to native tumors is paramount in these models. Given the intricate nature of cancer, any significant deviation from the original tumor could introduce unforeseen variables that might confound meaningful conclusions about tumor behavior.
While patient-derived xenografts (PDX) have emerged as a model system that allows for monitoring of tumor progression in vivo, some limitations question how well the developed tumor recapitulates human cancer biology4. For instance, melanoma TME contains tumor-infiltrating immune cells and stromal cells. When passaged as a xenograft, the non-cancer cells from the patient will be gradually lost, which can select adaptations that differ from those in patient tumors. While these limitations can be mitigated by performing limited passages of the PDX in vivo, the differences between the species can still pose a risk of genetic changes to the PDX that deviate from that of the native tumor5. Moreover, PDX models are typically established in immunocompromised mice to prevent rejection of human tissues. This limits the ability to study the role of the host immune system in cancer, especially in immunotherapeutic studies6. We and others have engrafted PDXs in "humanized" mice7,8. However, establishing PDX models in humanized mice is time-consuming and expensive. It can take several months to years to develop a single PDX model from a patient's tumor specimen, thus limiting the speed and scale of research. Despite these limitations, PDX models have been shown to accurately represent the biology of the TME, thus having been extensively used for the development of cancer therapeutics, personalized medicine, and immunotherapy, among others9.
An alternative 3D culture technique that has emerged is the patient-derived organoid (PDO), where a freshly resected patient tumor is cultured to create models that maintain phenotypic heterogeneity, histoarchitecture, and intercellular communications10. Its ability to closely resemble native tumors has been previously demonstrated where cultured organoids from non-muscle invasive (NMIBC) and muscle-invasive bladder cancer (MIBC) successfully recapitulated key aspects of the parental tumors11. Due to their potential and advantages over other models, PDOs offer a wide variety of tailored applications that other models lack, including but not limited to the study and development of immune checkpoint inhibitors along with cellular and targeted therapies. For example, we have previously used melanoma patient-derived organoids (MPDOs) to study the ability of tumor-infiltrating lymphocytes (TILs) to be expanded by IL-2 and anti-PD1 antibodies (αPD-1). Results showed that the TILs expanded by IL-2 and αPD-1 not only had a higher number but could also successfully infiltrate MPDOs and kill melanoma cells with higher efficiency than TILs expanded using other methods12. Other groups have shown similar results where the usage of anti-PD-1 and/or anti-PD-L1 leads to TILs remaining functional in PDOs, which subsequently leads to tumor killing13. Additionally, our group has also used MPDOs to assess γδ T-cell infiltration and killing ability14. PDOs' applicability spans a wide range of areas, including TIL expansion, cytotoxicity studies, and small molecule screening. All these applications highlight the potential and usability that this model confers. As such, we describe the detailed protocol for the culture of MPDOs.