Bioengineered Tumor Organoids

Abstract

Recent advances in biofabrication technologies, such as cell culture systems, and biomaterials have led to the development of three-dimensional (3D) cell culture platforms, such as tumor organoids. Tumor organoids are more physiologically accurate to the in vivo system, which they are intended to model, compared with traditional 2D cancer cell culture systems. Tumor organoids can mimic pathological and physical characteristics of tumors as well as maintain genetic stability of the cancer cells. Furthermore tumor organoids have advantage over animal models, being made from human cells and easily controlled in the laboratory to attain the desired tissue characteristics. In this section, we describe general tumor organoid technologies, the importance of the tumor microenvironment (TME) in model culture systems, and the use of tumor organoids in drug development and precision medicine. Organoid technologies continue to develop rapidly for applications in academic, clinical, and pharmaceutical settings.

Graphical Abstract

Components of the tumor microenvironment. The TME is composed of many different cell types, signaling factors, extracellular matrix components and vasculature that all significantly affect the tumor. These components must be considered when developing a tumor model system.

1.1. Introduction

Advances in tissue engineering and biomedical techniques over the past decade has given rise to the development of three-dimensional (3D) cell culture systems such as tumor organoids. Organoids are broadly described as small, cellular constructs that represent their in vivo counterparts to a much higher degree compared to conventional 2D methods. Despite having yielded extensive breakthroughs in cancer research, traditional 2D cell cultures have limitations in studying cancer disease and progression causing need for more complex in vitro models. Tumor organoids can be assembled from embryonic stem cells (ESCs), induced pluripotent stem cells (iPSCs), somatic stem cells, cancer cell lines, primary cancer cells and/or stromal cells in a specific 3D culture matrix; however, most models for precision medicine consist of a mixture of patient cancer and stromal cells[1–4]. The cell type(s) that is incorporated in the tumor organoid system is determined by the experiment at hand. For example, determining the effect of chemotherapy agents on normal colorectal organoids will require stem cell incorporation and crypt self-assembly whereas testing the metastatic nature of cancer in differing ECM environments will most likely involve controlled cancer cell lines with specific mutations. 2D monolayer cell cultures lacks the diversity of cell types, spatial organization, and overall 3D microenvironment seen in tumors and overcome in tumor organoids. Current 2D methods commonly look at the cancer cells in isolation whereas organoids allow the incorporation of many factors (gradients, cell-cell and cell-matrix interactions) that are present in vivo.

Tumor organoids occupy a range of distinct forms, each with their own strengths and weaknesses, and appropriateness for particular applications and can be combined when necessary. Collectively, the bioengineered platforms better mimic structure and cellular heterogeneity of in vivo tissue and are therefore more suitable for cancer research. In cancer, the tumor microenvironment (TME) is the dynamic space in and around the tumor. It contains a myriad of cell types, growth and paracrine factors, and structural components, each contributing to the progression of a tumor. The combination of these interactions determines the final fate of a cancer and has a massive impact on the severity of a patient’s disease. In this section, we will describe the clinical considerations for cancer treatment, the major components within the human TME, and how this can be modeled using tumor organoids.

1.2. Aspects of the tumor microenvironment

Tumors are more than merely a homogenous accumulation of cancer cells. They consist of a convoluted organization of cancerous and healthy cellular elements along with noncellular components. The tumor microenvironment includes the extracellular matrix (ECM), a variety of secreted factors, and different types of cells surrounding the central tumor tissue, has a significant role in regulating tumor growth, cancer malignancy and treatment response (Graph. Abstract) [5].

Tumor organoids have the capability to incorporate TME components to recapitulate some aspects of this complex dynamic depending on the cancer model. The ECM comprises of collagens, proteoglycans, laminins and fibronectin.[6] The ECM is crucial for both tissue function and maintenance. Due to its role on cellular behavior, an unregulated ECM can lead to dramatic consequences such as fibrosis and cancer [7]. ECM abnormality in tumors is an excess in degradation and/or deposition of any of the ECM components described and can result in cancer cell growth, survival, migration, and chemoresistance.[8–11] These changes in the ECM are modulated by the surrounding cells that have the ability to synthesis, degrade, align and crosslink the matrix and can regulate various cellular process such as gene expression, differentiation, and proliferation.[12–14] The 3D nature of tumor organoids enables them to reconstruct certain aspects of the ECM and determine involvement in cancer; however, it remains difficult to isolate the function of each ECM component on tumor progression due to the vast interdependent relationship each component has on one another.

There is a vast array of chemical signals, mainly growth factors (GF), found throughout the microenvironment, each with their own outcomes and signaling pathways. Altered levels of these different factors exert dramatic effects on cancer cell behavior. Of the many cytokines and GF, transforming growth factor β (TGF-β) is a major GF produced by cancer cells and the stromal environment. In a normal, homeostatic environment, TGF-β can induce cell cycle arrest; however, in many cancer types, an increase in TGF-β production is seen, suppressing immune and other reactive cells while inducing EMT in cancer cells [15]. TGF-β also stimulates CAFs to produce and release ECM, making it a key factor in tumor fibrosis and metastasis.

Major classes of cells present in the tumor microenvironment include cancer associated fibroblasts (CAFs), tumor endothelial cells (TECs), tumor associated macrophages (TAMs), and regulatory T cells (Tregs). Recent studies show that organoids can be used to gain insight on ECM interaction with stromal cells and identify function of stromal cells on cancer cells. One group developed organoids using pancreatic stellate cells and pancreatic cancer cells leading to the finding of a pancreatic CAF subtype that expresses less aSMA and secretes interleukins to support organoid growth.[16] Another group shows that IL-1 induces activation of signaling pathways that produce inflammatory CAFs and demonstrates that TGF-R antagonizes this process in adenocarcinomas.[17] This development can suggest multiple strategies to target these cells in vivo. Tumor organoid research has yet to answer larger questions concerning the TMEs’ involvement with cancer progression and drug resistance; however, these major discoveries is on the horizon and will likely emerge from tumor organoid models.

1.3. In vitro models of cancer

Major categories of tumor organoids and the methods by which they are biofabricated have become well established and include tumor spheroids, conventional organoids, device or membrane-based co-cultures (Figure 1). Tumor spheroids are created using U-bottom non-adherent culture plates or via the hanging drop method [18]. Tumor cell populations (homogenous or heterogenous) are added in suspension to the specified plates and allowed to self-assemble overtime (often just 3 days). As spheroids, tumor cells can create their own ECM to further resemble the in vivo microenvironment. The tumor spheroids can be used alone as 3D models of cancer or can be hydrogel encapsulated for additional complexity. Encapsulation allows for physical influences to impact tumor cell behavior and can facilitate the addition of stromal cells and ECM factors both soluble and in-soluble [19]. The use of hydrogels is not limited to spheroid encapsulation and can also be utilized to encapsulate stromal and tumor cell suspensions to study self-organization amongst other things. The use of membranes to create 3D tumor co-cultures systems additionally allow for the study stromal/tumor cell communication and impact on disease progression. In these systems, tumor cells and stromal cells are seeded opposite each other on a thin membrane and cultured together over time. These systems can be further integrated into tumor-on-a-chip devices which allow for the addition of flow into the model. Selection between each of the 3D tumor cell culture methods described ultimately relies on the goals and intended outcomes of the study but offer significant advantages over 2D culture [20].

Figure 1:

Three-dimensional organoid techniques that are used in cancer research. These bioengineered 3D organoid platforms better recapitulate structure and cellular heterogeneity of in vivo tumor tissue and are therefore more appropriate to model the desired tumor type.

1.4. In vitro models of different cancers

Models vary for cancer types and the purpose of the model being developed. For any single cancer type no single model is used and variation of models allows for a deeper understanding of the disease state, its progression, and treatment response. Below are examples of 3D models used for different cancer types showing the importance of considering the disease in three dimensions.

Breast

It is beneficial to study breast cancer in 3D, as the disease is associated with specific genetic mutations that behave in a more physiologically relevant manner in organoid models. For example, 15–20% of breast cancer patients carry HER2 mutations. Studies using 3D versus 2D models have shown that HER2 in 3D tumor spheroid cultures exhibits a substantial response to trastuzumab, a HER2-positive treatment, in comparison to 2D models [21]. The 3D models are shown to better mimic the in vivo microenvironment and more closely represent patient response to drug.

Colon

Oncogenic mutations in the epithelial cells lining the lumen of the intestine are the most common cause of colorectal cancer. In this unique environment, the tumor is located between two distinct microenvironments, the luminal space and the submucosa. Tumor cells will migrate towards the submucosa which offers nutrients but is also collagen-rich and primarily populated with smooth muscle cells (SMCs) [22]. 3D tumor spheroids have been encapsulated in collagen with SMCs to replicate this unique environment [22]. Unlike in 2D, the 3D model allows for re-organization of the stroma by stromal cells creating a relevant architecture and physiological cues for the tumor cells to be observed within.

Liver

Liver cancer is closely associated to cirrhosis, making the microenvironment a primary region of focus when understanding disease progression and developing treatments. Cirrhosis of the liver is caused by uncontrolled inflammation and ultimately scarring of the tissue, yielding a collagen and ECM protein rich environment for damaged hepatocytes. This altered microenvironment modulates cell activity and ultimately can be a contributing factor to cancer development and behavior. 3D models of liver cancer are able to incorporate these many cirrhotic components by utilizing collagen-based hydrogels and incorporating hepatic stellate cells to influence the microenvironment and tumor cell behavior [23].

Lung

Trials to develop 3D lung cancer organoids have resulted in models that contain many of the TME cell types to best recapitulate the in vivo tumor. It has been well established that the addition of both fibroblasts and endothelial cells make the model more physiologically representative. As lung cancer is the deadliest cancer in the US, researchers have utilized patient primary samples to create patient and tumor specific microenvironments for the study of disease progression and treatment response in 3D [24]. These methods are mixed culture and preservation of the various cell types within the tumor allow for acini structure development in 3D, which is important for phenotypic modeling [24].

Pancreatic

Similar to other types of cancer, pancreatic cancer treatment has been closely tied to stromal factors, making it important to consider the impact of stellate cells and fibroblasts on disease progression. Researchers have utilized co-culture models of pancreatic cancer to further understand these relationships [25]. Specifically, cancer spheroids have been added on top or within hydrogels containing stromal cells which have yielded physiologically and pharmacologically relevant structures such as lumens which have not been shown in stroma free cultures [25]. These improvements will hopefully allow better drug development to decrease the mortality rate in pancreatic cancer.

Prostate

Prostate cancer is unique in that it is often benign when discovered and left untreated until confirmed to be malignant. One developing treatment of prostate cancer has been the use of nanoparticles which allow for controlled, specific drug delivery to tumor regions, which reduces systemic chemotherapy response and need for invasive surgery. Studies have shown that 2D to in vivo nanoparticle deployment is often times unsuccessful and demands the use of 3D models for predicting efficacy of the treatments [26]. Utilizing custom scaffolds and stroma specific to the prostate microenvironment, researchers have been able to create nanoparticles translational from in vitro to in vivo [26].

1.5. Tumor organoids for clinical treatment predictions

Efforts to utilize patient derived organoids for the prediction of treatment sensitivity has become a driving component of their development [4]. These efforts have demonstrated the ability to predict genetic and proteomic sensitivities to treatments, often in a similar time frame as results from genetic screening. The most direct example of this testing is the screening of approved treatments to attempt to determine which may benefit the patient [27]. There are also efforts to reproduce rare forms of cancer in 3D to elucidate poorly understood cellular mechanisms [28]. Studies are usually aimed at testing large panels of approved and experimental compounds on previously genotyped patients in an attempt to discover novel treatment options [1]. Other studies focus on modeling tumor subtypes of the same cancer, with the ability to predict resistance patterns and suggest optimal treatments based the response of their organoid testing [29]. A more recent focus of the tumor organoid modeling has included intra-tumor sampling to create diverse sets of organoids derived from the same patient [30]. These organoids demonstrate the reality of tumor resistance in different sites of the same tumor. There are also projects to create organoid based biobanks to assist in precision medicine based studies of patient treatment [31]. Finally, implementation of microfabricated devices has allowed for the modeling of both treatment of tumor organoids under flow as well as the ability to prevent tumor cell migration [32]. Overall, there remains a high demand in the utilization of patient derived organoids in therapeutic determination, and many efforts are underway to implement them into clinical settings.

1.6. Tumor microenvironment and immune cell modeling

The ability to increase the functional state of a patient’s anti-tumor immune response by inhibition of normal immune checkpoint signaling, genetic engineering to create chimeric antigen receptor T-cells, or immune supplementation can illicit thorough cytotoxic activity with the potential to be more durable than traditional chemotherapeutics. Main targets for immunotherapy include, but are not limited to, natural killer (NK) cells and cytotoxic T-cells. These cell types operate through different pathways but are both central to anti-tumor immunity. Additionally, the efficacy of these therapies can be further compounded when used synergistically of other therapies such as inorganic photothermal nanoparticle agents that increase the presence of viable tumor antigen or precision medicine targeted therapeutics [33, 34].

We have recently created patient-specific immune-enhances tumor organoids by introducing immune cells from lymph nodes or the peripheral blood along with melanoma cancer cells [35]. We observed significant tumor cell killing in the response to immunotherapy, similar to clinical response in these patients. The creation of organoid models allows for the evaluation of factors that dictate immunotherapy response in a more controlled environment. The number of confounding variables in-vivo makes it much more difficult to assess the immunotherapy induce cytotoxicity and isolation of various factors that may be contributing to immunotherapy resistance. The ability to integrate patient tumor and immune cells into the organoid rather than cell lines can add to precision medicine approaches for patient evaluation. Patient cell inclusion has been performed with NK [36] and T-cells [37], T-cells [34, 38–44], NK cells [34, 38, 41], and B-cell populations [34, 41, 42]. Readouts from these systems can show if a patient will exhibit resistance to immune checkpoint inhibition or other immunotherapies, thereby informing the eligibility of the patient for immunotherapy. Examples include the creation of patient matched cytotoxic T-cells and tumor cells in order to display evidence of direct T-cell mediated killing in vitro using anti-PD-1 [34, 42–44], PD-L1 [34], CTL4 [43], and MICA/B and NKG2A [38] therapies.

1.7. Conclusions and future applications

Tumor organoids from cancer patients can be used to identify the ideal treatment for a specific patient since tumor organoids retain the genetic heterogeneity of the primary tumor as well as their TME, resulting in an attractive tool for precision medicine in oncology. There are many techniques that can be utilized in developing tumor organoids for drug discovery, precision medicine, genomic, or malignancy studies, to name a few; however, researchers all need to consider the TME and ECM when considering which method they wish to use. Models incorporating components of the TME may elucidate new mechanisms behind cancer control leading to novel therapeutics which target the support system of a cancer rather than the cancer itself. Additionally, where the production of organoid models had a relatively low throughput in the past, rapid developments in 3D bioprinting are making organoid models increasingly viable for drug screens [45]. Improvements in related 3D techniques are also allowing expansion in personalized medicine by comparing sequencing data and drug screening results do in vivo models [31]. These strategies could lead to a new landscape of cancer therapies which more directly target cancer cells and their stroma while sparing the body’s healthy tissues.

Abbreviations

3D

three-dimensional

ECM

extracellular matrix

TME

tumor microenvironment

CAF

cancer associated fibroblasts

GF

growth factor

TGF-β

transforming growth factor-beta

EMT

epithelial to mesenchymal

VEGF

vascular endothelial growth factor

EGF

epidermal growth factor

PDGF

platelet derived growth factor

FGF

fibroblast growth factor

TNF

tumor necrosis factor

IL

interleukin

TEC

tumor endothelial cells

TAM

tumor associated macrophages

Tregs

regulatory T cells

MMP

matrix metalloproteases

HER2

human epidermal growth factor receptor 2

SMC

smooth muscle cells

NK

natural killer