If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
1 Joana Rita Oliveira Faria Marques and Patricia González-Alva have equally contributed to the work.
Joana Rita Oliveira Faria Marques
Footnotes
1 Joana Rita Oliveira Faria Marques and Patricia González-Alva have equally contributed to the work.
Affiliations
Oral Biology and Biochemistry Research Group (GIBBO), Unidade de Investigação em Ciências Orais e Biomédicas (UICOB), Faculdade de Medicina Dentária, Universidade de Lisboa, Lisboa, Portugal
1 Joana Rita Oliveira Faria Marques and Patricia González-Alva have equally contributed to the work.
Patricia González-Alva
Footnotes
1 Joana Rita Oliveira Faria Marques and Patricia González-Alva have equally contributed to the work.
Affiliations
Tissue Bioengineering Laboratory, Postgraduate Studies and Research Division, Faculty of Dentistry, National Autonomous University of Mexico (UNAM), 04510, Mexico, CDMX, Mexico
Oral Biology and Biochemistry Research Group (GIBBO), Unidade de Investigação em Ciências Orais e Biomédicas (UICOB), Faculdade de Medicina Dentária, Universidade de Lisboa, Lisboa, Portugal
Faculty of Dentistry, National University of Singapore, SingaporeORCHIDS: Oral Care Health Innovations and Designs Singapore, National University of Singapore, Singapore
Cancer treatment development is a complex process, with tumor heterogeneity and inter-patient variations limiting the success of therapeutic intervention. Traditional two-dimensional cell culture has been used to study cancer metabolism, but it fails to capture physiologically relevant cell-cell and cell-environment interactions required to mimic tumor-specific architecture. Over the past three decades, research efforts in the field of 3D cancer model fabrication using tissue engineering have addressed this unmet need. The self-organized and scaffold-based model has shown potential to study the cancer microenvironment and eventually bridge the gap between 2D cell culture and animal models. Recently, three-dimensional (3D) bioprinting has emerged as an exciting and novel biofabrication strategy aimed at developing a 3D compartmentalized hierarchical organization with the precise positioning of biomolecules, including living cells. In this review, we discuss the advancements in 3D culture techniques for the fabrication of cancer models, as well as their benefits and limitations. We also highlight future directions associated with technological advances, detailed applicative research, patient compliance, and regulatory challenges to achieve a successful bed-to-bench transition.
Cancer is a major disease burden worldwide, with 19.3 million new cases and 10 million deaths in 2020, resulting in devastating health, economic consequences and is a major impediment to increasing life expectancy [
]. Although this is a significant and growing issue, the mechanisms of cancer initiation, development, and metastasis remain poorly understood. Moreover, the native cancer microenvironment (CME) is highly dynamic, with distinct extracellular matrix (ECM), cell–cell contact, and cell–matrix interactions that vary according to disease stage and regulate cancer growth, angiogenesis, aggression, invasion, and metastasis [
Monolayer culture of cells in two-dimensional culture (2D) and in vivo animal model is a traditional technique to study cancer-related microenvironment in cancer pathogenesis [
]. While the 2D model is cost-effective and simple to implement, it does not replicate the microenvironment and architecture (including cell-cell interactions and junctions) of a tumor in vivo [
]. Animal models can overcome these limitations; however, they are expensive, time-consuming, and often involve intricate surgeries. Most importantly, owing to the lack of human cells and their immune response, they fail to mimic human physiology, resulting in a failure to translate pre-clinical findings. In addition, several ethical concerns regarding animal welfare are also being raised; therefore, the use of animal models needs to be rationally and carefully managed [
]. To address the drawbacks of both above mentioned model intermediate three-dimensional (3D) models known as spheroids, organoids and gel embedding were developed to mimic cancer microenvironment [
Recent advances in tissue engineering have shown that an in vitro model can be developed to map realistic in vivo niches of cancerous tissues, allowing for a better understanding of the disease and the development of novel cancer treatment approaches [
]. These new platforms have been exploited to recapitulate the composition, architecture, and biomolecule presentation of native tissues ex vivo for in vivo implantation. Importantly, well-defined natural and synthetic scaffold-based tissue-engineered constructs (TECs) can mimic extracellular matrices and direct cell fate by promoting cell-matrix interactions [
]. The porous scaffold with different mechano-chemical properties can increase the culture efficiency and cell functions. With the advancement in 3D printing technologies to create perfusable networks and cell positions in a high-throughput manner to better mimic the cancer microenvironment [
]. This is especially important for recreating a functional cancer microenvironment that represents the 3D structure, stromal environment, and signaling milieu found in vivo.
In this three-part review, we discuss innovations in 3D cancer models to study the cancer microenvironment. In the first part, conventional approaches for generating cell aggregates, including spheroids (such as those formed through hanging-drop, overlay on agarose, and spinneret flasks) and organoids (which can be derived from patient-derived cancer organoids and organ-on-chip) are discussed. The second part evaluates how distinct biomaterials can be used to recreate the complex cancer microenvironment to study the disease. This includes recent work on 3D printing to render a microenvironment with controlled 3D architecture and biological features to develop functional models. Finally, we highlight the future direction with the utilization of biomaterial and 3D bioprinting technologies to develop biomimetic cancer models with effective translation to clinic.
2. Conventional approaches to fabricate in vitro 3D cancer models
Over the years, 3D culture models have been developed to understand cell-cell and cell-extracellular matrix interactions in the tumor microenvironment. Self-organized 3D models (Fig 1) are the most common methods for tissue engineering of cell-rich organotypic 3D tumor models [
]. These models have exhibited better immunomodulatory, proliferative, and activation abilities than 2D cultures, thus having a tremendous impact on studying the cancer microenvironment (Table 1).
Fig 1Schematic showing the structure and distinctions between the spheroid and organoid 3D culture for tumor microenvironment.
Spheroids are defined as self-assembled clusters of cell colonies that have some tissue-like organization, with cell-cell interactions dominating cell-substrate interactions [
]. These structures (> 500 μm) comprise three concentric layers of cells with distinct properties: internal necrotic layer due to the hypoxic core, intermediate quiescent layer, and external proliferative layer [
]. Tumor spheroids for drug screening can be generated by clonal expansion or through cell aggregation. For clonal expansion, single cells are immobilized in a hydrogel matrix, such as Matrigel polyethylene glycol (PEG) or fibrin, or by loading cells at a higher density into hydrogel microspheres [
]. For in vitro spheroid production in a simpler and faster manner, the most commonly used methods are based on inhibiting cell-surface interactions through a non-adherent environment: the hanging drop method, the use of low-attachment plates, overlay on agarose, and rotary cell culture [
]. In the following sections, these methods are described and discussed, along with their limitations and strengths.
2.1.1 Hanging drop (HD)
The hanging drop (HD) method has been used extensively based on its simple and high-throughput protocols that enable the screening of many molecules/conditions. Briefly, a small volume (20 μL to 50 μL) of cell suspension was pipetted onto a Petri dish or a multi-well special plate lid, which can be covered by an amphiphilic surfactant coating [
]. The lid was then inverted, and by the combined effects of gravity and surface tension (water-air interface), the cells were allowed to aggregate at the bottom of the droplet over time [
Due to its simplicity, low-cost and repeatability, this method is by far the most used in the literature for cancer spheroid generation. Several cell lines and primary cells modeling different cancer types have been proved to be able to generate spheroids using HD, including breast cancer (MDA-MB-231, BT-474, MCF-7, ES-D3) [
The cisplatin, 5-Fluorouracil, irinotecan, and gemcitabine treatment in resistant 2D and 3D Model Triple Negative Breast Cancer Cell line: ABCG2 expression data.
Anti-Cancer Agents Med Chem.2022; 22(Formerly Current Medicinal Chemistry-Anti-Cancer Agents): 371-377
Ovarian cancer stem cells and macrophages reciprocally interact through the WNT pathway to promote pro-tumoral and malignant phenotypes in 3D engineered microenvironments.
Evaluating the in vitro therapeutic effects of human amniotic mesenchymal stromal cells on MiaPaca2 pancreatic cancer cells using 2D and 3D cell culture model.
] among others. This method has also been proven useful for growing cancer stem cells (CSCs), since the number of cells loaded on each droplet is very small, and CSCs represent a small percentage of the entire population of tumor cells [
Ovarian cancer stem cells and macrophages reciprocally interact through the WNT pathway to promote pro-tumoral and malignant phenotypes in 3D engineered microenvironments.
]. It is also possible to perform co-culture of different cell populations representative of the tumor environment, such as embryoid bodies and tumor spheroids, with the aid of microfluidic devices [
]. In this latter approach, hydrogels are patterned using “press-on” techniques to create culture niches in with confined spheroid-generating geometries and cells are seeded to create spheroid-in-gel cultures of breast cancer [
]. However, medium exchange without disturbing cells can be challenging in these setups, and the use of commercial HD plates increases the cost of this technique [
]. In this method, a defined number of cells (ranging from 1000–20,000 ) were cultured in a liquid suspension on a non-adhesive surface, originally obtained by coating regular, flat-bottomed wells with a thin layer of agarose (1%). This prevents cell attachment to the surface of the well and drives cell aggregation. Other materials, such as scaffolds produced with poorly adhesive biomaterials such as polymethylsiloxane may be used [
]. Recently, hyaluronic acid (HA) has been proposed as an alternative coating material, since it is a non-adhesive polymer that has the ability of, unlike other materials, regulating cancer cell behavior via cell-HA signaling [
A study on polysialic acid as a biomaterial for cell culture applications.
J Biomed Mater Res A.2008; 85 (An Official Journal of The Society for Biomaterials, The Japanese Society for Biomaterials, and The Australian Society for Biomaterials and the Korean Society for Biomaterials): 1-13
]. Notably, in a study by Carvalho et al., by altering the HA concentration as well as the number of cells seeded in each well, the size, shape, and number of spheroids were controlled, and heterotypic spheroids with a co-culture of MCF-7 breast cancer cell line and normal human dermal fibroblasts (NHDF) were possible, simulating the interactions between cancer cells and cancer-associated fibroblasts [
] and head and neck carcinomas have been demonstrated to form spheroids with this technique.
Under optimal parameterization, spheroids obtained by LOT on 1–2% agarose may have homogeneous sizes, morphologies and stratification. Horizontal stirring between 150 and 200 rpm during the first six days of culture seemed to improve cell aggregation [
]. Some disadvantages of LOT on agarose are its rather limited adaptation to high-throughput screening, for example, the challenges in obtaining uniform coating of microtiter plates with agarose and the variability in spheroid formation across wells [
]. Another limitation is the fact that conventional microplates are prone to edge defects that may lead to evaporation of plate contents, particularly in the external wells, compromising drug testing [
2.1.3 Ultra-low attachment plates or non-adhesive culture plates (ULA)
The ultra-low attachment (ULA) plate method for generating spheroids is also a scaffold-free model that is based on the use of uncoated plates or plates coated with a hydrophilic polymer with a round or V-shaped bottom that enables forced floating and spheroid self-assembly. The principle is similar to that behind the LOT, and for that reason, the ULA may be considered a type of LOT. Spheroids generated using ULA are normally homogeneous and uniformly sized and may be generated with 500–20,000 cells per well. After 24 h , spheroid formation began, and visible aggregates were detected after a few days. Apart from the controlled size and shape of the spheroids, which is an advantage compared to other methods, high-throughput research and drug screening are enabled because ULA multi-well plates and automation are compatible with various testing platforms [
Comparative studies on the efficacy and suitability of HD and ULA methods for generating spheroids for chemoresistance assays have demonstrated that, while spheroid circularity is similar between methods, the size of the ULA spheroids is normally larger, suggesting that ULA may generate lower adhesive forces between cells in the spheroid, which may pose a limitation for some types of studies that require organoid transfer. Although the specific cells and cancer type to be modeled need to be considered, ULA has been consistently demonstrated to be more reliable and predictable than the HD method and therefore can be considered more adequate for generating spheroids for cytotoxicity assays [
Agitation-based methods are based on suspension cell culture systems that use a bioreactor to generate a constant dynamic fluid flow that results from the horizontal or vertical rotation of the vessel. This enables constant nutrient input and removal of waste products generated by long cultures while assuring an optimal balance with the negative effects of shear stress on cells. The continuous motion also ensures that the cells do not adhere to the container walls, but rather form cell-cell adhesions [
]. Rotating Cell Culture System™ (RCCS) developed by NASA and commercialized by Synthecon® Inc., spinner flasks are examples of such bioreactors that are also capable of creating microgravity conditions, which are potentially involved in mechanisms controlling cancer growth and function [
]. Spinner-flask bioreactors include a cell container holding the cell suspension and a stirring element that creates a constant motion of the suspension. In RCCS, the culture vessel itself rotates and creates a low shear force in the cell culture.
Agitation-based approaches may be considered when special conditions, such as normoxia or hypoxia, are to be simulated, as well as specific drug assays, including in vitro pharmacokinetic studies [
]. RCCS successfully integrates cellular colocalization, 3D stromal/epithelial/matrix interactions, and low shear forces, providing a quiescent environment for 3D spheroid cell culture with adequate mixing for mass transport [
]. However, spheroids produced by bioreactors are heterogeneous in shape, which means that in experiments where the size of spheroids needs to be controlled, manual selection of the spheroids generated and replating onto a culture plate is required [
]. Therefore, apart from requiring specialized and expensive equipment, this may be a time-consuming method because of the additional steps required if spheroid size and number control are required [
An organoid is a three-dimensional (3D) culture that has organ-like properties and must be able to mimic organ architecture to reproduce its functions [
]. Unlike spheroid cultures, in which the 3D culture is derived from monolayer cell lines, organoids describe 3D cultures established from tissue fragments [
]. In this sense, they include several cell types as seen in vivo, namely by culturing stem cells or cancerous tissue in matrices supplemented with specific molecules that are representative of native tissue [
]. This allows organoids to copy the genotype and phenotype of their corresponding tissues in vivo. Currently, several natural and synthetic materials are used for matrix formulation. The most commonly used matrices are based on collagen or Matrigel™ [
]. However, MatrigelTM has limitations, such as batch-to-batch variation and the nature of animal tumor-derived raw material composed of unidentified components [
]. Decellularized tissue matrices have been suggested as an alternative to conventional matrices because of their ability to better simulate the extracellular matrix (ECM). Another alternative approach is the use of hydrogels, for example, poly (ethylene glycol) . Cancer organoid modeling using a well-defined biomaterial can provide more typical disease phenotypes and pathophysiology [
Several approaches to perform organoid-like 3D cell culture have been described in the literature, including patient-derived organoids, engineered organoids, and organ-on-a-chip [
]. These models offer the possibility of patient-specific high-throughput drug screening to select the most effective therapeutic approach in personalized medicine. Additionally, PDCOs are a good option for investigating the properties and cellular heterogeneity of tumors because their main advantage is to reproduce the genetic and phenotypic heterogeneity found in the original tumors. So far, the majority of PDCOs described in literature are derived from epithelial cancers [
Conventional organoid culture mainly focuses on producing multicellular structures with either a single or multiple cell types. This structure is simple, disorganized, and limited in terms of tissue-level features [
]. This technique is used with organoid technology to overcome the disadvantages of conventional organoid culture, namely costs, and is very time consuming [
]. Although PDCO is well described in the literature, few reports on engineered organoid cancer models are available.
2.2.3 Organ-on-a-chip
The advancements in microfluidic technology, tissue engineering, and 3D printing have facilitated the development of organ-on-a-chip models for cancer research [
]. These models have emerged as powerful tools for studying human physiology and pathology at molecular, cellular, and tissue levels. To fabricate these models, 3D channels and chambers are patterned using flexible materials such as polymers or bioinks through soft lithography or 3D printing technologies [
]. These channels and chambers are then filled with fluids or hydrogels containing cells to mimic the 3D structure and forces of human tissue, with control over important physicochemical factors such as oxygen tension, interstitial pressure, and chemical gradients [
]. Several cancer-on-a-chip models have been developed for different primary and metastatic tumor types, including liver, lung, breast, prostate, colorectal, skin, metastatic liver, lung, bone, and brain cancers [
]. These cancer-on-a-chip models are capable of simulating key physiological, biophysical, and biochemical hallmarks observed in vivo, making them suitable for drug and immunotherapy screening under dynamic flow conditions [
Recent advances have focused on including lymphoid components in these models, thus enabling the study of dynamic mechanisms of tumor development, invasion, and resistance in interaction with surrounding tissues and the lymphatic system, including metastasis, hypoxia-state, and onco-immuno modeling in a cancer-type-specific way [
]. These immune-active platforms, although still in their early stages, have the potential to enable both drug and immunotherapy screening under dynamic flow, thus providing more meaningful results, with exciting perspectives and advances expected in the near future [
The current limitations of these models include the lack of standardization, technical challenges in production, and limitations in reproducibility and throughput. Cancer-on-a-chip uses mainly immortalized cell lines with limitations for in vivo extrapolation and unrealistic cell homogeneity. Future research should address these limitations by incorporating autogenous cells and relevant cell types in the tumor microenvironment, such as endothelial cells, cancer-associated fibroblasts, tumor cells, and organ/specific cell types [
Advances in biomaterial research in the area of scaffold development have been primed to meet the laboratory and clinical needs of numerous biomedical applications, including the rapidly growing area of engineering cancer microenvironments [
]. Scaffold-based cancer models provide an appropriate microenvironment where cells can reside and proliferate; moreover, they facilitate cell-cell and cell-ECM interactions in the native tumor microenvironment [
Three-dimensional collagen-based scaffold model to study the microenvironment and drug-resistance mechanisms of oropharyngeal squamous cell carcinomas.
]. These 3D scaffolds are produced from a wide range of materials and possess tunable porosities, permeabilities, surface chemistries, and mechanical characteristics [
]. In this section, we present scaffold-based 3D models fabricated using natural and synthetic assisted in replicating the tumor microenvironment and facilitating the testing of new cancer therapies (Table 2).
Table 2Scaffold-based 3D culture system for cancer cells.
Type of scaffolds
Cell type
Cell line
Outcome
Reference
Natural
Collagen
Breast cancer
MCF-7
Increase pro-angiogenic growth factors and matrix metalloproteinase transcription factors.
Depending on the application, scaffolding materials can be sourced from natural substances or biological sources, such as proteins commonly found in the ECM in vivo [
Three-dimensional collagen-based scaffold model to study the microenvironment and drug-resistance mechanisms of oropharyngeal squamous cell carcinomas.
]. In addition, biomaterials such as Matrigel™ and gelatin methacrylate (GelMA) have been explored to formulate 3D in vitro platforms, mainly in the form of hydrogels, because they can recreate ECM cues better than their free-scaffold counterparts [
]. An adequate material selection is crucial to the scaffold fabrication and must be compatible with the production technique to function effectively with in vitro models [
]. Natural polymeric scaffolds have low immunogenic properties and are popular options for in vitro culture models. However, their limitations include poor biomechanical properties, lot-to-lot variability, and difficult control of degradation rate and sterilization.
In addition, scaffold pore size and pore interconnectivity are essential features to consider in scaffold design because they determine the amount of oxygen and nutrients distributed over the scaffold [
]. Hence, different fabrication techniques can produce scaffolds with different porosities, pore sizes, shapes, and features. For example, freeze-drying method is an effective technique for fabricating scaffolds with a porosity of more than 90% and a pore size between 20 to 200 µm [
Three-dimensional collagen-based scaffold model to study the microenvironment and drug-resistance mechanisms of oropharyngeal squamous cell carcinomas.
] used freeze-drying to produce collagen hydrogels with a patient-derived primary culture of squamous cell carcinoma-papilloma virus-positive tumors, confirming the ability of collagen-based scaffolds to induce drug resistance. The authors pointed out that primary cultures conserved parts of stromal populations and different subclones of patient tumor tissue. The model could analyze drug resistance to standard chemotherapeutic agents used to treat oral cancer (5-fluorouracil and gemcitabine), corroborating an overexpression of epithelial-mesenchymal transition (EMT)-related genes and enhanced in vitro migration ability of the tumor. Although the former model is a step closer to personalized medicine, surgical samples are often difficult to obtain by laboratories and their management is complex [
Three-dimensional collagen-based scaffold model to study the microenvironment and drug-resistance mechanisms of oropharyngeal squamous cell carcinomas.
] cultured WSU-HN12 cells, a metastatic squamous cell carcinoma cell line, on cryogenic electrospun silk scaffolds, evaluated the sensitivity of paclitaxel and compared it with 2D culture conditions. Their results indicated that HN12 in 3D culture conditions were more resistant to the chemotherapeutic agent and showed similar features to in vivo conditions, including drug resistance and high Ki-67 expression. Head and neck cancer organoids established by modifying the CTOS method can be used to predict in vivo drug sensitivity by combining an epithelial cell sheet, MatrigelTM, and primary culture from head and neck cancer patients. The constructs were treated with cisplatin and docetaxel for one week. The 3D cultures showed increased drug resistance, and the results corresponded those of individual patient donors with recurrent head and neck cancer [
Hydrogels can capture vast quantities and retain a 3D structure; they can be easily analyzed to assess cell viability, proliferation, tumor formation, or onset of hypoxia [
]. Although hydrogel platforms can encapsulate cells and imitate native ECM, the arrangement of cancer cell upon incorporation is a crucial parameter to be evaluated [
]. For example, differential and random cellular arrangements in a hydrogel matrix may impact cell-cell adhesion, physiology, and drug resistance, consequently influencing therapeutic bioactivity [
] evaluated the in vitro maturation of MG-63 osteosarcoma spheroids and cell lines in GelMA and Matrigel™ hydrogels. Their results revealed that spheroids embedded in hydrogels exhibited an enhanced invasion phenotype and showed higher sensitivity to lorlatinib, a kinase inhibitor, when compared with their cell-laden counterparts.
In addition, physical and mechanical changes in the ECM, such as stiffness, are commonly overlooked by 2D culture models, which are heterogenic in tumors. For example, correlative maps obtained from breast tumor biopsies show that stiffness varies from 2 kPa in the core to 20 kPa in the periphery [
]. Also, tumor tissues are stiffer than their normal counterparts, for example., normal liver tissue has a stiffness of approximately 1.5–5 kPa, while fibrotic or tumoral tissue show 14–69 kPa stiffness [
Further improvements in 3D natural scaffolds have been achieved, usually by blending additional connective tissue components or biocompatibility compounds, especially growth factors, which could optimize scaffolding conditions and overcome the mechanical and biological limitations of the scaffold [
]. However, typical problems are associated with this type of scaffold, including the availability of materials that match the tissue of interest, tedious procedures, unwanted remnant proteins, and confusing signaling events [
]. The former is also true for synthetic polymeric scaffolds.
2.3.2 Synthetic scaffolds
Synthetic polymers are a popular choice for the fabrication of 3D scaffolds owing to their high biocompatibility, degradability, and resorption. Also, they can be used with standard processing techniques, such as melt electrospinning or air-jet spinning. Electrospinning techniques for highly porous 3D scaffold fabrication involve the use of high voltage or gas to create charges on a polymer solution ejected at a specified flow rate using a syringe pump [
]. Once the electrostatic forces exceed the surface tension of the fluid, the solution is pulled towards a grounded collector, and fiber mats are formed in the process [
Synthetic polymers can mimic the structural properties of the ECM; however, they fail to provide the biochemical signals needed for cell-ECM communication [
]. Also, most polymers have mechanical limitations, and they are often combined with inorganic materials that can enhance their properties, such as bioglasses or calcium phosphate [
Impact of 3-D printed PLA-and chitosan-based scaffolds on human monocyte/macrophage responses: unraveling the effect of 3-D structures on inflammation.
Moreover, the multiple combinations of natural and synthetic materials resulted in hybrid scaffolds that have helped advance cancer research in several ways, such that 2D culture models have not. Fig. 2 shows a schematic illustration of the scaffold-based 3D culture models. For example, several studies have used 3D Ca-alginate scaffolds as a cell culture platform for screening and testing the efficacy of anticancer drugs [
]. Also, malignant breast cancer cell lines (MCF-7 and JIMT-1) are fusiform, flat, and epithelioid in 2D culture plates; but grow small, round, and form aggregates and spheroids on 3D electrospun Poly(e–caprolactone) (PCL) mats Furthermore, breast cancer cell lines were compared with dermal fibroblasts, which grow in a spread pattern and elongated structures.
Fig 2Schematic illustration of scaffold-based 3D culture models. (A) natural hydrogels in which tumor cancer cells arrangement cannot replicate tumor heterogenicity; (B) modification of hydrogels with inorganic nanoparticles and growth factors, and spheroid cells have improved the understating of the tumor microenvironment; (C) Electrospinning techniques allow the fabrication of nanostructural three-dimensional scaffolds. The scaffolds mimic the extracellular matrix (ECM) and provide biochemical signals needed for cell-ECM communications.TC, Taylor cone; GACS; grounded aluminum collecting surface.
Consistent with this concept, prostate cancer cells grown on porous sphere-template poly (2 hydroxyethyl methacrylate) (PHEMA) show tumorigenic response, comparing to non-porous PHEMA the cells only respond in the presence of Matrigel™ [
Prostate cancer xenografts engineered from 3D precision-porous poly (2-hydroxyethyl methacrylate) hydrogels as models for tumorigenesis and dormancy escape.
]. It is essential to mention that functional peptides' integration actively serves as docking sites for interacting proteins. For example, CD44+/CD24- breast cancer cell culture in 3D self-assembling 16 residue peptide (RADA16) nanofiber scaffolds reverted their stem phenotype, reduced migration and invasion capabilities, and compared to type I collagen and Matrigel cultures [
Hybrid scaffold models have been employed for many cancer types. For example, MDA-MB31 gastric cancer cells were cultured in a scaffold composed of Poly Lactic-co-Glycolic Acid (PLGA) fibers and a GelMA hydrogel. The cells showed heterogeneous features; a fraction of cells showed a cancer stem cell-like phenotype, and other populations underwent EMT [
]. Similarly, Young et.al developed the platform “TRACER”, a collagen and cellulose tissue roll scaffold for radiation therapy screening using the FaDu cell and primary cancer-associated fibroblast stromal cells. The cell lines were cultured into a cellulose layer, and the first layer was formed by the cancer-associated fibroblast followed by a central collagen/agarose layer and a FaDu cell layer on the top. Then, cell layers (TRACER) were rolled onto an acrylic core, placed into a custom-made 50 ml tube, and then subjected to 5- or 10-Gray radial arc radiation. The results indicate that cancer-associated fibroblasts were not capable of offering radioprotective behavior. Although the results are not fully corroborated, the platforms showed a potential tool to study tumor response to radiation [
In summary, the choice of materials is crucial for successful scaffold fabrication, and natural polymeric scaffolds with low immunogenic properties are popular options for in vitro culture models, though they have some limitations. The scaffold's pore size and pore interconnectivity are also essential features to consider in scaffold design. Hydrogels can capture vast quantities and retain a 3D structure, but cancer cell arrangement upon incorporation is a crucial parameter to be evaluated. Furthermore, synthetic polymers are a popular choice for the fabrication of 3D scaffolds due to their high biocompatibility, degradability, and resorption, and can be used with standard processing techniques.
3. 3D printing methods for cancer microenvironment
The main disadvantage of existing cancer models is the inability of these 3D tumor models to replicate identical human physiological conditions and related functioning. Biofabrication of native tumor-like microenvironments with comparable cellular and ECM compositions and architectures can significantly enhance the investigation of cancer molecular and biological processes. 3D bioprinting is an emerging biofabrication technology in many areas, including tissue engineering and cancer research. It allows customization of precise spatial arrangements of multiple materials, cell types, and diverse extracellular matrix (ECM) components within a construct to simulate the cancer microenvironment (Fig 3) [
]. Current 3D printed cancer models have shown in vivo tumor behaviors such as high growth rate, aggressive invasiveness, angiogenesis, metastasis and high resistance to anticancer drugs (Table 2) [
]. In a general workflow, a 3D model of the desired construct is designed and generated using computer-aided design (CAD) software. The 3D model file can then be transferred into slicing software, where parameters such as layer thickness and print path can be determined. Upon slicing, the software generates a G-code file to relay various print settings to the 3D printer. During printing, materials are deposited layer-by-layer and subsequently into the designed 3D model [
Different 3D printing techniques are being utilized to biofabricate 3D in vitro cancer models that may be used as preclinical drug testing platforms and serve as bridges between in vivo and human testing (Fig. 4). The choice of 3D printing technique depends on the type of biomaterial used to mimic the tumor microenvironment, desired biomechanical properties, and complexity of the model. The most common 3D printing techniques used for 3D cancer models are stereolithography, inkjet-based, extrusion-based, direct light-patterning-based printing, and laser-assisted printing [
Stereolithographic (SLA) printing uses light of specific wavelengths to polymerize liquid polymers into thin layers. Ultraviolet (UV) light sources are commonly used to trigger photocrosslinking of materials and cure liquids in a rapid and consistent manner. SLA offers high print resolution with great structural integrity, complexity, and cell viability [
]. Therefore, limited by photopolymers, SLA has a narrow range of material choices. Photopolymers such as polyethylene glycol dimethylacrylate (PEGDMA), polyethylene glycol diacrylate (PEGDA), or GelMA with photo-crosslinkers are used in SLA bioprinting to serve as scaffolds for cell attachment or as a printed tissue [
SLA in cancer models utilizes its high print resolution, where refined structures with high complexity can be recreated. In a breast cancer model, nano-ink with hydroxyapatite nanoparticles in the hydrogel serves as a bioink to mimic the role of the bone-specific environment in breast cancer metastasis [
]. In an ovarian cancer model, a mini-scaffold composed of polycaprolactone (PCL) was fabricated and placed into 96-well plates for culturing. GelMA with cells was loaded onto the scaffold for investigation of the cancer microenvironment [
A 96-well microplate bioreactor platform supporting individual dual perfusion and high-throughput assessment of simple or biofabricated 3D tissue models.
A 96-well microplate bioreactor platform supporting individual dual perfusion and high-throughput assessment of simple or biofabricated 3D tissue models.
]. Viscous gel-like bioinks are often used to ensure stable structure. With simple modular design, conventional extrusion-based printers incorporate multiple print heads to construct various materials at once [
]. However, the resolution of the construct depended on the microextrusion nozzle head gage. Layering by pattern-drawing motion results in a slower printing process, and with shear stress created by passing cells through the nozzle head, cell viability can be compromised [
]. Parameters including material concentration, nozzle gage size, and dispensing pressure should be optimized for each cell-material combination for the best viability and ink flow [
]. Popular materials in EBB include collagen, gelatin, alginate, and GelMA, which serve as scaffolds for cell attachment and precise shape formation.
As the most economical method of bioprinting, EBB is the most widely used technology for various cancer models. 3D printed mini-brain model for glioblastoma (GBM) was fabricated by including GBM-associated macrophages and GBM multiforme cells. It was comparable to GBM samples from over 150 patients [
]. In 2014, an EBB 3D pancreatic tumor model was filed for patency. It has been reported to recreate oncogenic signaling processes by bioprinting with stromal bioinks. Microenvironment establishment is not achievable in 2D or spheroid cultures [
Inspired by conventional inkjet printers, inkjet bioprinters deposit materials such as ink sprays. Cells or materials are dispensed in microdroplets of ∼50 µm diameter with high throughput in a non-contact and precise drop-on-demand manner [
]. Low-viscosity liquids are frequently used as bioinks. Multiple micro nozzle print heads were installed to spray a large volume of droplets efficiently with the flexibility to incorporate multiple materials simultaneously. Contactless and high-speed dispensing ensures better cell viability than extrusion-based bioprinting, despite sharing similar concerns regarding shear stresses at nozzle heads. The disadvantage of inkjet bioprinting is the lack of precise deposition of each ink droplet. The material viscosity must be optimized to prevent nozzle head clogging [11] Dilute agarose, collagen, alginate, GelMA, and Matrigel™ were used for inkjet printing. The spraying nature of inkjet printing enables large-tissue construction with the specific incorporation of different cell types.
Drop-on-demand printing with multiple cell types is often adopted for inkjet bioprinting. A GBM model comprising endothelial, stromal, and cancer cells loaded with collagen type-1 was co-printed using multiple inkjet nozzles. A vascular system construct was created to successfully mimic the spread of neuroblastoma [
]. The high-throughput nature of inkjet bioprinting was used in an ovarian coculture model. A reproducible model with a controlled cell density was fabricated. Human ovarian cancer cell OVCAR-5 and fibroblast co-culture micropatterning on Matrigel™ were modeled for systematic investigation [
Digital light processing (DLP) 3D printing technology uses a projected light source to cure the entire layer and enables improved speed for the fabrication of complex 3D geometries and precise control over material properties [
]. To investigate the influence of HCC growth and invasion on 3D matrix stiffness, a hepatocellular carcinoma (HCC) model was created using HepG2 cell-laden liver dECM-based, Col-I, and GelMA bioinks with variable matrix stiffness. Imaging of stromal invasion behavior from nodules with diseased liver stiffness was made possible by 3D bioprinted hexagonal nodules of varying stiffness. Interestingly, stiffer scaffold demonstrated more invasive and migratory capacity at both the genetic and phenotypic levels [
]. Another team studied the effects of ECM stiffness on glioblastoma using bioinks consisting of glycidyl methacrylate hyaluronic acid (stiff) and gelatin methacrylate (soft). The stiffer model demonstrated extensive infiltration of tumor cells and increased resistance to temozolomide. Meanwhile, the soft models enrich the classical subtype and promote expansive cell proliferation [
]. In addition, DLP demonstrated controllable spatial architecture within a conventional 96-well cell culture, which is capable of high-throughput cancer models for preclinical drug screening [
Laser-assisted printing utilizes laser energy to pattern cell-laden bioinks in a three-dimensional spatial arrangement with the aid of computer-aided design and manufacturing. Pancreatic ductal adenocarcinoma was bioprinted using (GelMA), composed of both acinar and ductal cells, to mimic the initial stages of adenocarcinoma. The model displayed high cell viability and the expression of pancreatic cancer-specific markers [
]. Similarly, multicellular tumor spheroids were processed using core–shell structures with cancer cells, and stem cells showed high cell viability. Furthermore, the influence of tumor spheroid size on transferrin internalization, a typical ligand for targeted treatment, shows increased spatial heterogeneity in large aggregates [
Some of the 3D bioprinted cancer models are summarized in Table 3, which shows the various bioprinting techniques that have been used to mimic tumor microenvironment features. In summary, biofabrication of 3D in vitro cancer models using 3D printing technologies offers a promising platform for investigating cancer's molecular and biological processes, as well as pre-clinical drug testing. 3D bioprinting enables the precise spatial arrangement of multiple materials, cell types, and diverse extracellular matrix (ECM) components within a construct to simulate cancer microenvironments with comparable cellular, ECM composition, and architecture to native tumors. The choice of printing technique depends on the type of biomaterials used to mimic tumor microenvironment, desired biomechanical property, and complexity of the model.
Table 33DBP studies of the various cancer tumor microenvironment.
Cancer Type
Cell Type
Bioink/ Material
3D Bioprinting Method
Outcome
Ref
Glioblastoma
Glioma cell line U118 and endothelial cells
Collagen or dECM hydrogel
EBB
Patient specific ex-vivo model of glioblastoma with anatomical organization is created. Oxygen gradient leading to hypoxia environment is mimicked.
Patient samples were incorporated on chip for drug combination treatment and analysis.
Human glioma stem cell GSC23 (shell); Human glioma cells U118 (core)
Sodium Alginate
Coaxial EBB
Drug resistance comparison between core-shell group and core-only group was done. Core-shell group shows higher protein and mRNA expression of treatment resistance markers. Making a potential model for drug test
3D construct with customizable shapes and sizes were printed. Cells cultured in 3D constructs were found to have higher invasiveness and drug resistance than 2D culture.
Precise micropatterning of co-cultured cell droplets were loaded onto microfluidic chips. Drug diffusion gradient over the distance is studied, cell viability can be assessed from the dosage gradient
A novel approach for precisely controlled multiple cell patterning in microfluidic chips by inkjet printing and the detection of drug metabolism and diffusion.
Cell-laden hydeogel scaffolds are created. mRNA expression for cell migration, angiogenesis and proliferation were greatly enhanced in 3D printed model
Investigated the role of Glioblastoma-associated macrophages (GAMs) in cancer development through 3D printed co-culture. The model was able to quantify cell migration in 3D spatial arrangement, potentially an effective way to study immunemodulatory and chemotherapy responses
Human primary umbilical cord-derived mesenchymal stromal cells (UC-MSC, referred to as MSCs), HUVEC, and human bone marrow-derived epithelial-neuroblastoma immortalized cells (SH-SY5Y)
Agarose and type-I collagen
DBB
Stable bioprintable collagen hydrogel was optimized without contraction. Cancer cells formed Homer Wright-Like Rosettes in the model, recreating the histological pattern similar to high-risk patients’ samples. The model is suitable for studying cancer development and serves as the first line in therapy medicine test.
Immortalized non-tumorigenic human breast epithelial cell line MCF-12A; Breast carcinoma cell lines MCF-7 and MDA-MB-468
Neutralized rat tail collagen I; Human collagen I VitroCol; Growth-factor reduced Matrigel (Geltrex)
EBB
Reproduceable mammary epithelial organoids were generated. At day 21, a large duct-like structure was observed, potentially generating larger tissue to investigate breast carcinogenesis
Normal breast epithelial cells HMLE; Twist-transformed cells HMLET
Poly (ethylene glycol) diacrylate
LBB
With different stiffness of substrates, behavior of HMLE and HMLET are similar in 2D culture for growth displacement, velocity, and straightness. 3D culture, however, shown a significantly higher migration of HMLET over stiff substrate, displaying a more promising cancer migration process over 2D culture.
Immortalized non-tumorigenic human breast epithelial cell line, MCF-12A, and the breast carcinoma cell lines MCF-7 and MDA-MB-468
Rat tail collagen
EBB
The study shows that a low-cost bioprinting platform can be used to increase tumoroid formation in 3D collagen gels, generate precise tumoroid arrays, and mimic in vivo findings.
Breast cancer cells BT474; Human perinatal foreskin fibroblasts BJ and human adult dermal fibroblasts HDF for Integration-free human induced pluripotent stem cells (iPSCs) generation
Poly(ethylene glycol) diacrylate (PEGDA)
LBB
Spheroid simulating cancer hypoxic core were printed in a high throughput manner. With both low and high cell seeding density, both sample sets were able to reach a similar final diameter.
Breast epithelial cell lines MCF10A, MCF10A-NeuN, MDA-MB-231, and MCF-7
Matrigel, Gelatin, Collagen
Coaxial EBB
All three types of bioinks were successfully printed and cell viability was compared. Spheroid co-culture was printed with different cell combinations for drug tests to show different interactions
Breast cancer cell lines of distinct subtypes, luminal (MCF-7), basal like (HCC1143), HER2 amplified (SKBR3), and claudin low (MDA-MB-231)
Alginate and gelatin
EBB
Therapeutic efficacy and genetic expression of microenvironment of 3D printed breast cancer models was investigated. In addition, the research went further to construct pancreatic cancer model with primary patient derived tumor cells for drug treatment.
IMR-90 fibroblast cells and MDA-MB-231 cancer cells
Alginate and gelatin
EBB
Cell migration in a customized 3D printed construct was studied. Recruitment of fibroblasts by the cancer cells was visualized over 30 days, The model could be further explored for therapeutic tests or metastasis studies.
Primary human bone marrow cells MSCs; Human adenocarcinoma BrCa cell line MDA-MB-231
GelMA and nanocrystalline hydroxyapatite nHA
LBB
Model for breast cancer metastasis is constructed. Co-culture of MSC and BrCa under 3D condition shows enhance proliferation of BrCa and inhibited MSC growth.
Holmes B, Zhu W, Zhang LG. Development of a novel 3D bioprinted in vitro nano bone model for breast cancer bone metastasis study. MRS Online Proceedings Library (OPL) 2014, 1724.
Human adenocarcinoma BrCa cell line MDA-MB-231; Human fetal osteoblast cell line hFOB
PEG hydrogel and nHA
SLA
Artificial bone scaffold was fabricated. 3D culture was able to demonstrate the osteoblast recruitment upon cancer development, simulating the metastatic behavior.
Human ovarian cancer cell line OVCAR-5; Normal human fibroblasts MRC-5; Human primary umbilical vein endothelial cells HUVEC
Matrigel
DBB
This work is in line with (79). 3D Model of micrometastatic OvCa was fabricated. Co-migration of OVCAR-5 and MRC-5 were observed. When fibroblast is printed closer toOVCAR-5, tumor size dramatically increased over 14 days.
Human ovarian cancer cell line (SKOV3) and human foreskin-derived fibroblasts (HFF)
GelMA; Gelatin-norbornene (gel-NOR); PCL
SLA
SKOV3-HFF co-culture was captured in PCL scaffold and cultured in 96-well plate. With bioreactor, reproduceable and high throughput screening was achieved.
A 96-well microplate bioreactor platform supporting individual dual perfusion and high-throughput assessment of simple or biofabricated 3D tissue models.
Special 3D hepatic triculture models were printed. Tri-cultured cells reorganization and realignment were observed. Genetic expression indicated high maturity of cells in the model
HOS, 143B and U2-OS (human osteosarcoma cell lines)
GelMA, HAMA
EBB
RNA-Seq, qPCR and chemotherapeutic drug cytotoxicity evaluation have screened across 2D culture, spheroid culture and 3D bioprinted model culture. Autophagy response was found to be prominent in bioprinted model
Through cell proliferation analysis, 3D bioprinted model shows higher cell density and stronger resistance to drug treatment, simulating a more accurate cancer microenvironment.
4. Future research directions and concluding remarks
The development of effective cancer treatments is a complex process due to tumor heterogeneity and inter-patient variations, which limits the success of therapeutic interventions. Traditional two-dimensional cell culture has been used to study cancer metabolism, but it fails to capture physiologically relevant cell-cell and cell-environment interactions required to mimic the tumor-specific architecture. The fabrication of 3D culture reproducible models is a significant advancement in cancer research. The use of 3D cancer models has shown potential in addressing this unmet need by capturing the cancer microenvironment, including cell-cell and cell-environment interactions, and bridging the gap between 2D cell culture and animal models. The recent emergence of 3D bioprinting as a novel biofabrication strategy has enabled the development of compartmentalized hierarchical organizations with precise positioning of biomolecules, including living cells, for the fabrication of 3D cancer models.
Despite the benefits of 3D cancer models, there are limitations associated with technological advances, detailed applicative research, patient compliance, and regulatory challenges that must be addressed to achieve a successful bed-to-bench transition. The absence of a standardized approach to characterize the properties of bioink and printed materials in bioprinting poses difficulties in comparing models across the literature. The integration of advanced imaging techniques and computational modeling can further enhance the development of 3D cancer models. Additionally, the use of patient-derived cells and biomaterials with tuned stiffness in the fabrication of 3D cancer models could enable a more personalized approach for cancer treatment. It will be crucial to establish proper culture and research techniques, supported by comprehensive data on cancer genomics and proteomics, which will be essential in the optimization of relevant 3D models for predicting therapeutic outcomes for specific cancers. Smart polymers can be used to develop programmable tumor microenvironments by changing their volume or other properties in response to stimuli, such as pH, to mimic different stages of cancer. Taken together, the progress made in 3D models for studying the cancer microenvironment is significant in terms of understanding cancer development and drug response, and has the potential to provide groundbreaking solutions to replace animal models, ultimately transforming the healthcare industry.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The cisplatin, 5-Fluorouracil, irinotecan, and gemcitabine treatment in resistant 2D and 3D Model Triple Negative Breast Cancer Cell line: ABCG2 expression data.
Anti-Cancer Agents Med Chem.2022; 22(Formerly Current Medicinal Chemistry-Anti-Cancer Agents): 371-377
Ovarian cancer stem cells and macrophages reciprocally interact through the WNT pathway to promote pro-tumoral and malignant phenotypes in 3D engineered microenvironments.
Evaluating the in vitro therapeutic effects of human amniotic mesenchymal stromal cells on MiaPaca2 pancreatic cancer cells using 2D and 3D cell culture model.
A study on polysialic acid as a biomaterial for cell culture applications.
J Biomed Mater Res A.2008; 85 (An Official Journal of The Society for Biomaterials, The Japanese Society for Biomaterials, and The Australian Society for Biomaterials and the Korean Society for Biomaterials): 1-13
Three-dimensional collagen-based scaffold model to study the microenvironment and drug-resistance mechanisms of oropharyngeal squamous cell carcinomas.
Impact of 3-D printed PLA-and chitosan-based scaffolds on human monocyte/macrophage responses: unraveling the effect of 3-D structures on inflammation.
Prostate cancer xenografts engineered from 3D precision-porous poly (2-hydroxyethyl methacrylate) hydrogels as models for tumorigenesis and dormancy escape.
A 96-well microplate bioreactor platform supporting individual dual perfusion and high-throughput assessment of simple or biofabricated 3D tissue models.
A novel approach for precisely controlled multiple cell patterning in microfluidic chips by inkjet printing and the detection of drug metabolism and diffusion.
00026-2&id=gr1.jpg){kind=link}
00026-2&id=gr2.jpg){kind=link}
00026-2&id=gr3.jpg){kind=link}
00026-2&id=gr4.jpg){kind=link}