Novel Three-Dimensional Cultures of Patient-Derived Cancer and Tumor Immune Cells

Neal JT, Li X, Zhu J, Giangarra V, et al. Organoid Modeling of the Tumor Immune Microenvironment. Cell 2018;7:1972-1988.e16. Organoids (alternatively referred to as enteroids) are 3-dimensional cultures of tissue that have transformed basic and translational research in many organ systems and cancer subtypes, particularly in the gastrointestinal tract and liver. However, commonly used organoid models lack a functional immune cell component, which limits their application for studies of the tumor stroma (composed of fibroblasts and immune cells) or immunotherapies that activate a patient’s immune system to induce an antitumor response. In this study, Neal et al report the development of a method to culture tumor epithelial cells and immune cells together in 3-dimensional culture. The same group previously reported an air–liquid interface (ALI) model of culturing mouse-derived intestine, stomach, and pancreas (Nat Med 2009;6:701–706) and tissues derived from mouse models of gastrointestinal cancer (Nat Med 2014;7:769–777). In this system, intestinal fragments containing both epithelial and mesenchymal/stromal components are cultured in a collagen matrix and media containing specific growth factors. In the current study, the authors report successful culture of patient-derived organoids using the ALI system from surgically resected primary and metastatic cancers representing a wide range of cancer subtypes, including colorectal, pancreas, esophagus, and stomach. These in vitro cultures should be distinguished from patient-derived xenografts in which patient tumor tissue is engrafted into immunodeficient mice. In a major advance for cancer modeling, these new organoid models preserve the immune cell diversity of the original patient tumor, including T, B, macrophage, and natural killer cells. Finally, organoids from cancer subtypes known to respond to programmed death 1 (PD-1) inhibition (eg, non-small cell lung cancer and melanoma) demonstrated tumor lymphocyte expansion and activation in concordance with inhibition immune cell PD-1 expression. Gastrointestinal cancers such as colorectal and hepatocellular cancers are major causes of cancer mortality. Preclinical and translational studies have commonly been performed in human cancer cell lines, cell line transplants into immunodeficient mice (ie, xenografts), or mice that have been engineered to carry cancer-associated mutations (eg, loss of function mutations in the Apc tumor suppressor gene). Organoids are 3-dimensional structures that reproduce histologic and genetic features of tissues. In the most commonly used approach, intestinal crypts or cancer cells are suspended in a 3-dimensional collagen matrix and cultured with supportive growth factors (Gastroenterology 2011;141:1762–1772; Cell 2015;161:933–945). Cancer organoids have been generated for a wide range of tumor subtypes, including colorectal, esophageal, and liver. Clinical trials of patients with gastrointestinal cancer and their corresponding cancer organoids have shown that patient-derived organoids closely match the original tumor in response to anticancer agents (Science 2018;359:920–926). Furthermore, co-culture of cancer organoids and peripheral blood lymphocytes has demonstrated T-cell–mediated killing of tumor cells (Cell 2018;174:1586–1598). Neal et al provide tools that may serve as important platforms for preclinical, translational, and therapeutic discovery in oncology. These models offer several advances compared with the currently available systems. First, the authors demonstrate that patient-derived organoids recapitulate the histologic and genetic features of the original patient cancer. Second, they show that a wide range of cancer subtypes, including gastrointestinal cancers, can be successfully cultured using this method. Third, organoids grown using the ALI approach reproduce cancer-associated fibroblasts, which are believed to play an important role in cancer progression and metastasis. Fourth, these organoids exhibit a T-cell repertoire that closely mirrors the T-cell signature found in the originating patient cancers. Finally, at least a subset of cultured tumors demonstrate loss of PD-1 expression, T-cell expansion, and cytotoxicity in response to nivolumab, a PD-1 inhibitor that is used clinically. The immune cells surrounding tumors are increasingly recognized to play essential roles in cancer formation and progression. Immune checkpoint blockade with PD-1 and CTLA inhibitors have revolutionized cancer treatment for a range of cancers by activating an antitumor T-cell response. Therapeutic response to immunotherapy correlates with tumor mutational burden and varies widely by tumor subtype. Notably, only the small proportion of colorectal cancers that are microsatellite unstable (ie, microsatellite unstable–high) respond to immunotherapy. Thus, there is great interest in understanding mechanisms of how cancers evade immune surveillance and why some cancers respond to immune checkpoint inhibitors while others do not. Currently, immunotherapy research is limited by the lack of models to study patient-derived cancers together with the same patient’s immune microenvironment. Immunotherapy has advanced based on research in mice, whose immune systems differ from humans in significant ways. Thus, the major impact of the current study is that scientists have a way to study how the tumor stroma for different tumor types and patients impacts tumorigenesis. The most exciting potential application of this research (which will require further study) is to study a patient’s response to immunotherapy (alone or in combination with other treatments) in the culture dish, to predict how the patient may respond. The authors were able to assess the response of patient-derived organoids to checkpoint inhibitors within 7 days of tissue harvest, which may be a clinically relevant time frame. Finally, there is also interest in targeting cancers by inhibiting tumor-associated fibroblasts, particularly in pancreatic cancers that are highly infiltrated with stromal cells. Although this circumstance was not a focus of the current study, the presence of fibroblasts in the organoid tissues opens this possibility. There are several potential limitations to this study and open questions that are relevant to research in gastrointestinal cancers. First, it is unclear how the culture conditions for the tumor stroma in ALI-based tumor organoids differ from more commonly used tumor organoids in collagen matrix (eg, Matrigel). The authors speculate that culture of tissue chunks without the use of enzymatic digestion, as well as improved oxygenation, may contribute to successful long-term culture of stromal fibroblast and immune cells using the ALI method. Further research may reveal methods to culture tumor fibroblasts, immune cells, and cancer cells in a collagen matrix, without the need for the ALI. Second, ALI-based tumor organoids may be challenging to derive and maintain: the authors report a 73% success rate of culture at 1 month across tumor subtypes and 80% success with cryopreservation with the most rapidly growing lines, with loss of tissue architecture and stroma after long-term passage. The authors note that culture of tumor-associated stroma was not possible beyond 60 days, even with supplementation with IL-2. Third, more work is needed to define the variability of fibroblasts and immune cell composition over culture time and passaging, which appears to vary across cancer subtypes. Finally, immune profiling was performed in human lung cancers, renal cell cancers, and melanomas, which were most likely selected owing to their known response to checkpoint blockade. It will be important to perform further studies to understand how these models may be applicable to studying the immune environment in gastrointestinal cancers such as colorectal cancer. Major goals for colorectal cancer research include understanding why microsatellite stable colorectal cancers respond poorly to immunotherapy and studying how to increase neoantigen presentation in these tumors to augment efficacy of immunotherapy. The study by Neal et al provides an exciting new platform for addressing these important questions, and many others, in the era of cancer immunotherapy.