Introduction

According to 2020 data, approximately 19.3 million new cancer cases were reported globally, resulting in around 10 million related deaths.1 With the high incidence and mortality rates of cancer posing a serious threat to human health, there is an urgent demand for the development of more effective anticancer drugs. Nonetheless, the screening of such drugs remains a highly challenging and burdensome task due to these statistics.2 Typically, tumor models are used to assess their anticancer effects, providing highly intuitive results.

The construction of tumor models has evolved from 2D to animal models and then to 3D models such as organoids, representing a gradual development. Although simple to use and economical, 2D cell culture models do not effectively replicate the in vivo environment. Animal models are widely used currently, offering a more complex biological environment. However, the major issue lies in the inevitable physiological differences between animals and humans, casting doubt on the reliability of the results.

Due to advancements in tissue engineering, bioprinting technology has emerged as a significant tool in the field of oncology. This technique involves the use of 3D printing to assemble cells, biomaterials, and growth factors in a precise and quantitative manner. The technology can create three-dimensional tumor tissue models that simulate the microenvironment of tumor occurrence and development. These models provide a more realistic representation of cell-cell and cell-matrix interactions, closely resembling the biological characteristics of in vivo tumors. This aids in the in-depth study of tumor growth, progression, metastasis, and other mechanisms, enabling rapid, high-throughput assessment of drug efficacy and toxicity, thereby facilitating personalized medicine. Today, 3D tumor spheroids, organoids, and tumor chips are successful products cultivated by the era. The combination of these tumor models with bioprinting technology holds promise to provide faster and more accurate solutions for screening tumor treatment methods, ultimately benefiting millions of cancer patients. In simple terms, biological inks are synthesized from cells, bio-polymers, and other auxiliary components, followed by the use of appropriate bioprinting techniques to fabricate tumor models (Figure 1). This article provides an overview of the latest developments in bioprinting and 3D tumor models from a biomedical perspective, such as tumor spheroids, organoids, and tumor chips. We begin with a brief summary of the current state of size-based bioprinting technology, followed by a description of the latest advancements in closely related bioinks. Finally, we discuss tumor models in detail, outlining the challenges and prospects of bioprinting technology in the construction of tumor models.

Overview of Bioprinting Technology

Bioprinting technology, as a crucial component of the two fundamental elements of bioprinting (bioprinting technology and bioinks), refers to the technique of creating three-dimensional objects by digitally controlled, continuous layer deposition of living cells and tissues as materials. From the earliest inkjet printing to the present, bioprinting technology has evolved into various forms and categories. The common ones we encounter include Inkjet bioprinting, Acoustic droplet ejection bioprinting, Laser-assisted bioprinting, Microfilament bioprinting, Stereolithography bioprinting, and Digital light processing (Figure 2). The classification can be based on the presence or absence of a nozzle, which includes nozzle-based jetting or extrusion printing, as well as nozzle-free, open-platform printing. To succinctly summarize these printing technologies, we have categorized existing techniques based on the additive manufacturing characteristics of bioprinting, using the following classification: zero-dimensional droplet-based, one-dimensional filament-based, and two-dimensional plane-based (Table 1).3–5

Table 1 Advantages and Disadvantages of Bioprinting Technology

Zero-Dimensional Microdrop Bioprinting

This type of technology uses controllable cell-containing micro-drops as the structural forming unit, mainly including inkjet, acoustic, laser-assisted, and laser-induced transfer.6,15–17 During the cell printing process, all four methods mentioned above have potential sources of cell damage, including thermal effects, tension from bubble rupture, instantaneous impact force during droplet ejection, and dehydration due to droplet evaporation. Overall, through effective process optimization, the micro-droplet-based printing process can achieve a cell survival rate of over 85%. In terms of shaping, the prominent advantages of these micro-droplet-based methods are high forming accuracy and fast printing speed, while the main drawback is the low ink viscosity, making droplet fixation difficult and thus challenging for forming larger complex three-dimensional structures.

Bioprinting using inkjet technology was first developed in 2003 and officially patented in 2006. This method is based on the traditional 2D inkjet printer, where the ink is replaced with bioink and the platform is switched to X-Y-Z control, enabling the 3D fabrication of biologically relevant objects. Bioink is deposited by controlled deformation of the nozzle’s internal space as it flows through, driven by piezoelectric or digitally controlled thermal forces, causing the controlled volume to generate droplets through the nozzle. These droplets distribute the liquid onto a collection platform, forming layers in the z-direction to ultimately create a 3D object. Inkjet bioprinting has advantages such as a simple system and relatively low cost.18 Multiple print heads can work in parallel, achieving high-resolution (≈30 μm) and fast manufacturing.19 However, there are limitations to inkjet bioprinting, such as the relatively low cell density (<106 cells/mL) used for bioprinting and the ability to print only bioinks with viscosities in the range of 3.5–12 mPa s. Christensen et al successfully printed branched vascular structures using inkjet bioprinting by controlling and uniformizing the channel diameter.20

Bioprinters currently available with nozzles have limitations in printing resolution, as they are expected to cause clogging when the size is less than 100 µm.21 Additionally, decreasing the nozzle diameter increases shear stress during the printing process. When subjected to critical shear stress, mechanical damage to the printed cells may lead to cell death.22 In order to address these drawbacks, Jentsch et al7 introduced a novel 3D bioprinting technique based on the principle of acoustic droplet ejection (ADE). Ultrasonic signals emitted by transducers pass through a water tank, then through the bottom of a small coupling reservoir filled with a hydrogel suspension containing cells, ultimately producing ejected droplets containing cells. These droplets move upwards and adhere to a movable construction platform. This method eliminates the need for a nozzle, reducing critical shear stress. Numerical simulations indicate that the maximum shear stress in the ADE process is 2.7 times lower than that of a 150 µm microvalve nozzle.

Laser-assisted bioprinting utilizes laser direct writing or laser induction forward transfer. The first application of laser-assisted printing to bioprinting was documented by Guillotin et al23 in 2010, outlining bioprinting using micro-droplet arrays, setting the stage for its application in biomedicine. The system employing this technique typically comprises three layers from top to bottom: an energy absorption layer, a donor layer, and a bioink layer. The pivotal component in this system is the donor layer, which reacts to the applied laser beam. At the top of this donor layer is an energy absorption layer (such as titanium or gold), and at the bottom, a thin layer of bioink is suspended for bioprinting. When a selective laser beam is directed at the energy absorption layer, the corresponding area of the donor layer beneath it vaporizes, generating a high-pressure bubble at this interface. This pressure propels the bioink, leading to droplet deposition on the collection platform. By controlling the z-level of the collector, a 3D structure is eventually created. The advantage of laser-assisted bioprinting is that cells are not directly subjected to high shear stress. In fact, there is no contact between the dispenser and the bioink during the bioprinting process. Therefore, this bioprinting method can achieve a relatively high cell viability (>95%) and can deposit highly viscous substances (1–300 mPa.s). However, the main drawback of this system is its high cost, primarily due to the necessity for a high-resolution and high-intensity laser diode. Additionally, concerns have been raised that using a laser beam to impact the basal layer may result in the release of metal particles, which could compromise the process’s cell compatibility.8

One-Dimensional Microfilament Bioprinting

This technology involves the utilization of one-dimensional gel microfilaments as the basic building block for bioprinting, specifically in reference to microextrusion-based 3D bioprinting. Microextrusion-based bioprinting stands as one of the earliest developed and most widely applied bioprinting techniques because of its capacity to construct intricate large-scale structures and to accommodate the creation of natural or synthetic high molecular weight polymer materials with varied crosslinking mechanisms and rheological properties.9 Unlike droplet-based printing methods, microextrusion-based methods do not suffer from issues such as bubble bursting or instantaneous impact that can cause cell damage. The main source of cellular injury from this method is the shearing action on the cells during the extrusion process, especially when the ink material has a high viscosity,10,11 as shown by data demonstrating a 13% decrease in cell viability after printing.24

The fundamental mechanism of using extrusion for 3D bioprinting in the field of biology was first described by Iwan et al in 2002.25 Similar to the classic fused deposition modeling (FDM) process, this technique involves the extrusion of materials into filaments, which are then layered to construct a structure. Two main fluid distribution systems have shown high performance in bioprinting applications: pneumatic and mechanical drives.3 Pneumatic systems can be valveless or valve-based, with valveless systems being more straightforward to fabricate and thus being the most commonly used in bioprinting.26 Valve-based systems offer various advantages in terms of pressure and pulse frequency control, enabling precise material deposition and high-resolution bioprinting. Mechanical systems are primarily driven by pistons or screws. Piston-based systems enable direct control over the deposition of bioink onto the platform. Screw-driven systems are better suited for high-viscosity bioink due to their improved spatial control, but they may pose a risk to cells due to the increased pressure drop at the nozzle outlet.26

In comparison to previously mentioned platforms, extrusion-based bioprinting offers several benefits, such as the capability to deposit high-viscosity bioink (30 mPa s to > 6×107 mPa.s) and higher cell density (> 108 cells/mL to cell spheroids).9,27 Extrusion-based bioprinting systems have the capacity to continuously and consistently extrude bioink, making them more suitable for most applications due to the preservation of the printed tissue structure. As cells undergo significant shear stress during the extrusion process, the typical cell viability after extrusion-based bioprinting ranges around 40–80%.9

The driving force solves the problem of the dynamic source of microfilament extrusion, while the formation of microfilaments and other process treatments directly affect the quality of the final printed product. In the formation of microfilaments and forming structures, the ink materials generally undergo gelation or crosslinking. Pre-crosslinking treatment is used to partially crosslink the ink before printing to facilitate direct extrusion into filaments. Temperature control treatment is used for various thermosensitive ink materials and adopts targeted heating or cooling treatments to achieve the desired filament formation effect.28,29 Post-crosslinking treatment takes on various forms, with the general principle that the extruded ink is crosslinked into filaments at the nozzle tail end, such as the simultaneous injection of ink and crosslinking agents, coaxial injection, atomization treatment, and direct printing in a crosslinking agent solution, etc., with these different forms of treatment widely used in sodium alginate hydrogel printing.30–32 As a new approach to enhance the resolution of current bioprinting platforms, electrospinning technology makes use of the electrostatic properties of microfilaments. This method employs rapid stretching of charged polymer solutions or melt jetting, together with solvent evaporation and solidification, and then collects the filaments as fiber mats.33 Following the application of voltage between the syringe and collector, the solution ejected from the metal needle transforms into a charged jet, which is then drawn towards the collector. As the solvent evaporates during the journey from the syringe to the collector, the diameter of the jet reduces significantly, leading to the production of numerous fibers deposited on the metal collector. Recent advancements in electrospinning technology have enabled the production of finer-diameter fibers, making it a promising method for enhancing the resolution of existing bioprinting platforms. Typically, the equipment comprises an X-Y-Z robot workbench and a standard electrospinning device, incorporating a solution or molten polymer extrusion system equipped with an injection pump or pneumatic regulator, as well as a high-voltage power supply.34–36 Electrospinning allows for the control of bioprinting in a continuous or intermittent manner. One of the primary challenges in printing electrospun fibers is addressing the erratic oscillations of charged jets, which can lead to unstable fiber and structure formation.37

Two-Dimensional Planar Bioprinting

Stereolithography (SLA) and Digital Light Processing (DLP) are two commonly used photopolymerization techniques in the field of bioprinting. Both utilize a light source to cure liquid photosensitive resins and can complete the bioprinting process on a two-dimensional plane. However, there are some differences in their technical implementation and application characteristics. Each technology has its own advantages, and the choice between SLA and DLP depends on the specific application requirements. SLA is more suitable for manufacturing parts that require extremely high precision, while DLP is ideal for the rapid printing of small, delicate structures. In the field of bioprinting, both of these technologies are employed in the fabrication of biocompatible scaffolds and tissue models.

The origins of stereolithography for additive manufacturing can be dated to 1984, when Charles W. Hull outlined the process of creating 3D objects by selectively converting the physical state of fluid in a resin layer into a solid through layer-by-layer photopolymerization.38 Stereolithography technology utilizes a laser beam to scan liquid photosensitive resin point by point, constructing each layer one point at a time. This method allows for very precise control over the curing process and is suitable for the fabrication of high-precision components. However, due to the point-by-point curing nature, the printing speed of SLA is generally slower and it is not ideal for mass production. In the medical field, the earliest use of stereolithography involved the production of models for cranial surgery, enabling researchers to generate highly precise and detailed skull models.39 This technology refers to stereolithography for cell printing, which is suitable for photocrosslinking bioinks. During printing, a photocrosslinking light source is projected onto the bioinks through a digital projector, selectively solidifying the bioinks layer by layer through photopolymerization, with the process controlled by a movable workstation along the Z-axis. By projecting 2D images onto the bioink reservoir, stereolithography allows for the creation of intricate 3D structures without the need for the printing head to move along the x-y direction. This characteristic leads to a faster bioprinting speed, potentially surpassing that of droplet and filament-based methods.12 Within the area exposed to light, the ink selectively gels based on the digital model, leading to the layered stacking necessary for creating a three-dimensional structure.40,41 As one of the earliest invented 3D printing technologies, it is widely applied in the field of rapid prototyping. High cell viability (>85%) is achieved by bioprinters through the selective photocrosslinking of bioinks, which avoids subjecting the cells to shear stress.9 For direct cell printing, apart from ink wastage, difficulty in structural detachment, and oxygen inhibition, it may also cause significant light stimulation and photoinitiator toxicity to the cells. Additionally, the limitation of ink material types is a drawback of this technology. One drawback of this system is the requirement for the liquid to be transparent and for limited scattering; otherwise, the material will not be uniformly penetrated by the light, resulting in uneven crosslinking.

Stereolithography technology has been improved by integrating a digital micromirror device (DMD) into the laser optical path, enabling the entire layer to be cured at once without the need for point-by-point scanning. The DMD, composed of many micro-mirrors, can rapidly switch between on and off states. Based on the loaded image information, the DMD precisely controls the reflection of light through the optical system and irradiates the photosensitive resin surface in a specific pattern, achieving rapid curing of the target area. DLP technology uses a digital micromirror array (DMD) to project light patterns, thereby curing photosensitive resins. This method allows for precise control of the curing area by changing the projected light patterns, enabling the printing of complex structures. With a high resolution and faster printing speed due to the simultaneous curing of the entire layer, DLP is suitable for batch production. Moreover, DLP technology can operate over a broader range of light wavelengths, increasing the flexibility of material selection and the diversity of applications.13 These advantages make DLP technology widely applicable in bioprinting and other high-precision 3D printing fields.

Tang et al42 used digital light processing (DLP) technology for rapid bioprinting, constructing a mixed cell model containing patient-derived glioblastoma stem cells (GSCs), macrophages, astrocytes, and neural stem cells (NSCs). For the first time, they simulated the complex microenvironment of glioblastoma in vitro, including the infiltration of immune cells and the interaction between tumor cells and normal brain tissue. This demonstrated the versatility of the bioprinted model in simulating the tumor microenvironment, including the invasiveness of tumor cells, drug resistance, and immune interactions. Mei et al43 bioprinted a non-small cell lung cancer (NSCLC) model using DLP technology and simulated the human vascular system within the model. They precisely constructed a three-dimensional lung cancer model with central blood vessel channels using a bioink composed of methacrylated gelatin (GelMA). This model not only included simulated blood vessel structures but also featured an inlet and outlet to connect to a perfusion system that could supply nutrients and oxygen through a peristaltic pump, simulating the flow of blood in the human body. In the study by Ma et al,14 live/dead cell staining tests conducted 1 and 7 days after printing with DLP technology showed that the cell viability was greater than 95%, indicating that the DLP printing process did not harm the cell survival rate.

Bioink Design

Bioinks are a special type of ink used in the field of bioprinting, composed of multiple components including cells, cell scaffolds, and other auxiliary ingredients44,45 (Table 2). The following are common components found in bioinks: Cells: Bioinks typically contain a variety of cells, such as stem cells, fibroblasts, chondrocytes, etc. The selection of cells depends on the specific application purpose and the type of tissue to be constructed. Cell scaffolds: These are usually biodegradable bio-polymers, such as gelatin, alginate, etc., which are often present in bioinks. These bio-polymers provide structural support for cell adhesion and assist in forming the desired three-dimensional structures during the bioprinting process. Additionally, to promote cell growth and differentiation, other components such as growth factors and cytokines, including fibroblast growth factor (FGF) and vascular endothelial growth factor (VEGF), may be incorporated into the bioink. These elements mimic natural biological signals, enhancing cell viability and tissue regeneration. It is worth noting that the specific composition of bioinks can be customized and improved according to different research and application needs. Researchers can choose the appropriate combination of ingredients to prepare bioinks based on the target tissue structure, cell characteristics, and requirements of the printing equipment. Moreover, the formulations and preparation methods of bioinks are continuously evolving and improving to enhance printing performance and cell viability.

Table 2 The Composition of Bioink

Cells in Bioink

In the construction of tumor models using bioinks, there are various options for cell sources based on the research or experimental objectives. Established cell lines, commonly referred to as cell lines, are a popular choice for cell sources because they can be cultured in vitro for extended periods and are readily available. Examples of commonly used tumor cell lines include HeLa cells and K7M2 cells, which originate from different types of tumors and can be selected for cultivation and use according to specific needs. Patient-derived primary cells: To more closely simulate the actual tumor environment, researchers may choose to use primary cells obtained from tumor patients (typically mature, terminally differentiated cell types). These cells can be isolated after resection and undergo limited in vitro expansion before being utilized to construct tumor models. This cell source retains the biological characteristics of the original tumor, enabling the creation of personalized tumor models. However, due to the complexity of their tissue origin, there may be variations in cell types and biological properties. Stem cells, with their potential for self-renewal and multilineage differentiation, can also serve as a cell source for constructing tumor models. Stem cells, with their capacity for self-renewal and differentiation, can mimic the behavior of tumor-initiating cells, which are often responsible for tumor initiation, progression, and recurrence. They can also generate cell populations containing various stages of differentiation, thereby better simulating the heterogeneity within tumors.57 As an alternative solution to address the limited availability of cells, mesenchymal stem cells from adult bone marrow are among the most commonly used stem cells.46 They have the ability to differentiate into various tissue types and can be expanded and differentiated into mesenchymal tissues such as bone, cartilage, tendon, fat, and bone marrow stroma both in vitro and in vivo.58 Furthermore, stem cells derived from perinatal or amniotic sources can also be expanded and differentiated into multiple cell lineages, including adipocytes, osteoblasts, myogenic cells, endothelial cells, neuronal cells, and hepatocytes, thus serving as an alternative source of multipotent cells.59

For constructing tumor models using bioinks, there are several options for cell sources depending on the research or experimental objectives. Established cell lines, known as cell lines, are a common choice for cell sources because they can be cultured in vitro for extended periods and are easily accessible. Commonly used tumor cell lines include HeLa cells, K7M2 cells, etc., derived from different types of tumors, and can be selected for cultivation and use based on specific needs. Patient-derived primary cells: To closely simulate the real tumor environment, researchers may opt to use primary cells obtained from tumor patients (usually mature, terminally differentiated cell types). These cells can be isolated after resection and undergo limited in vitro expansion before being used to construct tumor models. This cell source retains the biological characteristics of the original tumor, achieving personalized tumor models; however, due to the complexity of tissue origin, there may be variations in cell types and biological properties. Stem cells, which have the potential for self-renewal and multipotent differentiation, can also serve as a cell source for constructing tumor models. Stem cells, with their ability to self-renew and differentiate, can be used to simulate the behavior of tumor-initiating cells, which are typically responsible for the initiation, progression, and recurrence of tumors. They can also generate cell populations containing different stages of differentiation, thereby better simulating the heterogeneity within tumors.57 As an alternative solution to the limited availability of cells, mesenchymal stem cells from adult bone marrow are the most commonly used stem cells.46 They can differentiate into various tissue types and can be expanded and differentiated into mesenchymal tissues such as bone, cartilage, tendon, fat, and bone marrow stroma in vitro or in vivo.58 Additionally, stem cells derived from perinatal or amniotic sources can also be expanded and differentiated into multiple cell lines, including adipocytes, osteoblasts, myogenic cells, endothelial cells, neuronal cells, and hepatocytes, thus serving as an alternative source of multipotent cells.59

Cell Scaffolds in Bioink

Tissues are not composed of a single type of cell. The function of the liver primarily relies on hepatocytes, but other essential cells such as cholangiocytes, hepatic stellate cells, and Kupffer cells are also crucial. In the brain, billions of neurons are surrounded by astrocytes, which provide necessary nutrients to neurons and contribute to immune modulation.60,61 Human tissues are typically composed of cells, extracellular matrix (ECM), various growth factors and bioactive molecules, all existing in a complex and coordinated manner.60,62,63. The extracellular matrix (ECM) is a complex network of macromolecules synthesized and secreted by various tissues and cells in the body, including fibroblasts, mesenchymal cells, and epithelial cells. These molecules are distributed and accumulate in the cell surface and interstitial substances. The ECM forms the cellular scaffold of tissues and organs, providing a microenvironment for cell survival and growth. It significantly influences cell adhesion, differentiation, proliferation, migration, and functional expression. Natural and synthetic hydrogels have the capacity to mimic the microenvironment of human tissues, with good biocompatibility that promotes the exchange of nutrients and waste with cells and can partially restore ECM functions, making them ideal materials for cell scaffolds.64

Over the past few decades, several natural and synthetic hydrogels have shown great potential as cell scaffolds in bioprinting. Alginate, silk fibroin, gelatin, collagen, chitosan, agarose, and hyaluronic acid have all been extensively studied as bio-inks for cell scaffolds.47 These cell scaffolds exhibit a range of sensitive properties, including ionic sensitivity, photopolymerization, thermosensitivity, enzymatic sensitivity, and pH sensitivity, allowing simple gels to form solid 3D structures before, during, or after the bioprinting process. Alginate is a commonly used bio-ink component in medical and bioprinting applications due to its adhesive properties and straightforward calcium chloride crosslinking mechanism, making it an attractive material for bioprinting.48,49 However, its chemical structure often hinders cell adhesion, requiring the addition of other natural polymers, such as gelatin-based materials, collagen, or silk fibroin, to induce cell adhesion and bioactivity.51 Gelatin and methacrylated gelatin (GelMA) are also widely used in bioprinting due to their excellent cell compatibility and mechanical properties.50 However, the use of GelMA requires photoinitiators for crosslinking and solidification, which may be toxic to cells at high concentrations.51 Similarly, hyaluronic acid and chitosan, due to their biodegradability and good biocompatibility, can be modified into printable cell scaffolds for the construction of cell-loaded structures, providing tunable performance and mechanical properties.52,65

At the macroscopic level, synthetic materials offer superior physical properties, enabling the manipulation of the mechanical, structural, and geometric characteristics of bioinks, including elastic modulus, tensile strength, porosity, and alignment. In contrast, natural materials provide an appropriate structural and biochemical environment for cell encapsulation and positioning. We need a material that can combine the advantages of both types of materials while closely replicating the structure and environment of human tissues. A successful bioprinting material should possess the following characteristics: printability, biocompatibility, mechanical properties, and shape and structure. (1) Printability: The material should be suitable for deposition during the printing process, with appropriate viscosity, shear properties, response transition times, and suitable sol-gel transition stimuli. Extrusion-based techniques typically require materials with higher viscosity and shear thinning properties, while inkjet-based techniques require low viscosity materials with very short sol-gel response and transition times. (2) Biocompatibility: In the field of bioprinting, bioinks must have appropriate degradability, support cell adhesion, and not cause severe immune reactions or toxicity.66 (3) Mechanical properties: The mechanical properties of bioinks in terms of stiffness, elasticity, and strength should be consistent with the target tissue’s mechanical properties. (4) Shape and structure: The printed structures should closely resemble the shape and structure of natural tissues.

Compared to hydrogels with more uniform compositions, the ECM in the body is secreted and composed of many cells. Specifically, cells release specific molecules and proteins in the appropriate environment, which aggregate to form the ECM. This includes collagen, fibronectin, proteoglycans, and bone matrix proteins, among others.67 During the ECM secretion process, cells selectively release specific ECM molecules based on their environment and functional needs. These molecules form a three-dimensional structure on the cell surface, providing support and influencing the environment for cell morphology, differentiation, and function. The ECM also regulates biological processes such as cell signaling, cell-cell interactions, and cell migration. Notably, the composition of ECM components secreted by different types of cells may vary depending on the cell type and the tissue or organ in which they are located. For example, in bone tissue, osteocytes secrete an ECM rich in collagen and bone matrix proteins, while in connective tissue, fibroblasts secrete an ECM rich in collagen. Various sources, such as dermis, bladder, small intestine, mesothelium, pericardium, heart valves, and different species, can provide ECM.68 Nagao et al69 identified methods for extracting hydrogels from human renal cortex. Compared to normal renal tissue, this renal ECM (K-ECM) hydrogel contains a large amount of natural matrix proteins, providing niches that simulate the natural microenvironment of the respective bodies (Nagao et al, 2016). Bi et al recently discovered that liver-derived ECM (L-ECM) obtained from pigs can induce the differentiation of rat mesenchymal stem cells into hepatocytes.70 In the construction of tumor models, it is crucial to retain non-cellular components such as growth factors, which play an important role in cell growth, differentiation, migration, and tissue regeneration. For example, fibroblast growth factor (FGF) can promote cell proliferation and neovascularization, which is essential for tissue repair and regeneration, and the state of FGF can affect the rate of tumor progression.54 Transforming growth factor-β (TGF-β) is responsible for regulating cell proliferation, differentiation, and migration, and is also involved in tissue remodeling and fibrosis processes, which can slow down tumor progression under certain conditions.55 Vascular endothelial growth factor (VEGF) specifically promotes the proliferation of vascular endothelial cells and the formation of new blood vessels, which is crucial for vascularization in tissue engineering. In the study by Jain et al,56 inhibition of VEGF led to a reduction in vascular permeability, thereby increasing the delivery of oxygen and therapeutic drugs to tumors. These growth factors and cytokines are usually present in bioinks in an active form and can directly or indirectly influence cell behavior and tissue development. By precisely controlling the release and concentration of these factors, bioinks can provide an optimized growth environment for specific tissue engineering applications.

Decellularized extracellular matrix (dECM) is a natural scaffold produced by removing cellular components from tissues or organs while retaining their three-dimensional structure and some native fibrous components, such as collagen fibers. Due to its natural biocompatibility, tissue specificity, and ease of fabrication into usable bioinks, dECM is gaining increasing attention.71 Simple hydrogel preparation does not possess these advantages. dECM derived from various tissues has been shown to serve as a bioink for 3D bioprinting. Jang and colleagues53 created innovative tissue-specific dECM bioinks using adipose, cartilage, and cardiac tissues. Specific cells from each tissue type were able to proliferate for at least 14 days in these tissue-specific bioinks. These cell-laden dECM bioinks can be deposited through extrusion from a nozzle and then gelate at physiological temperatures to maintain the generated 3D structures. These bioinks immediately gelate at 15°C and form a stable crosslinked hydrogel within 30 minutes at 37°C, without the need for any stringent crosslinking conditions or gelation additives. Overall, dECM is a promising biomaterial that can serve as an essential component of bioinks, providing an optimized microenvironment for the growth of specific tissues, including but not limited to cartilage, cardiac, adipose, and liver tissues. Bioinks derived from ECM offer unparalleled biocompatibility, providing a favorable environment for tissue regeneration. The gel-like natural biomaterials are conducive to encapsulating cells and growth factors. In cell printing technology, various types of printers are typically used to directly print the corresponding structures with bioinks. However, their limited mechanical properties pose a challenge to their contribution to the physical properties of tissues, making the 3D shape of printed structures composed solely of hydrogels very challenging.72 Nevertheless, bioinks can be crosslinked and modified chemically to alter their mechanical strength or degradation time.73

Bioinks generally need to meet various criteria. For instance, the viscosity must be low enough to be extruded through a printing head while preventing high shear rates within the loaded units, but it should also be high enough to maintain the 3D shape after extrusion. These requirements have led to a significant focus on shear-thinning hydrogels in many bioprinting applications. Shear-thinning hydrogels exhibit low viscosity under high shear stress and return to their original viscosity once the shear stress is removed.3,74 The addition of supplements like graphene nanosheets can significantly enhance the tensile strength of soft hydrogels.75 The incorporation of a small amount of graphene (0.1–0.3 wt%) into chitosan can increase the elastic modulus of chitosan by about 200%. In vitro colorimetric and cell adhesion test results indicate that chitosan with graphene still possesses good biocompatibility. Beyond the passive effects on loaded cells through mechanical forces, bioinks should also directly influence cell survival and behavior. The biological properties and biocompatibility of bioinks play a crucial role in ensuring the generation of the desired tissue. The performance of an ideal bioactive hydrogel scaffold should reflect the structural and biological characteristics of natural tissue ECM. Currently, hydrogels still face significant challenges in replicating the diverse biological functions and mechanical properties of ECM.

Overview of Tumor Model Construction

From a temporal perspective, the development of tumor models can be broadly categorized into three types: two-dimensional cell culture models, animal models, and models that can replicate the tumor microenvironment.

(1) Two-dimensional cell culture models (Time span: early 20th century to present): The earliest tumor models involved the growth of cancer cells in two-dimensional cell culture dishes. These models were used to study the fundamental processes of cell growth, metabolism, and the effects of drugs on cells. However, 2D cell culture models have limited capabilities in simulating the complex tumor microenvironment within the human body, and struggle to replicate the diverse phenotypes,76 gene and protein expressions,77 and in vivo drug responses78 of cancer cells under 2D conditions.

(2) Animal models (Time span: 1950s to present): To gain a comprehensive understanding of tumor development and assess the effectiveness of new therapeutic methods, scientists began using animal models, such as mice and rats. Even today, animal models remain the benchmark for drug screening and cancer research, providing a more complex and realistic physiological environment for studying tumor growth, metastasis, and treatment responses. Animal models offer insights into more complex biological environments, whole-system responses, and in vivo drug metabolism and side effects. However, these models still fail to faithfully replicate the essential elements of the human tumor microenvironment (TME). The extensive use of animals in laboratory research has also raised ethical concerns, emphasizing the need for alternative methods to more accurately and clinically relevantly recreate the in vivo TME. Specifically, they only provide the mouse physiological microenvironment for human tumor cells, lacking key features of natural human organ and biological system functions. Moreover, the compromised immune system in mice may affect cancer progression, reducing their reliability in drug testing.79 To address these challenges, humanized mouse models have recently been developed, utilizing gene-editing technologies to study the interaction between cancer and the human immune system.80

(3) The tumor model replicates the tumor microenvironment (Time span: 1970s to present): The extracellular matrix (ECM) provides structural support for cells. In the context of the tumor microenvironment, the ECM plays a crucial role in controlling the behavior and interactions of tumor cells. As a component of the human tumor microenvironment (TME), the ECM interacts with tumor cells, immune cells, and other factors, influencing tumor formation and progression. The TME is a multifaceted environment, including tumor cells and surrounding non-cellular components, such as immune cells, blood vessels, fibroblasts, and matrix molecules. To accurately create corresponding tumor models in vitro, we must strive to replicate the tumor microenvironment as much as possible. To better simulate the real growth environment of tumors, scientists have begun to culture tumor cells in a way that captures their essence. Among these are the most representative 3D tumor spheroids, organoids, and tumor chips (Table 3), which are three different but interconnected technologies, each with unique applications yet overlapping and complementary relationships.81 These tumor models allow for better simulation of the interactions between tumor cells within tissues and multicellular tissue structures, providing more physiologically relevant models that can offer a more realistic and accurate representation of the tumor microenvironment, aiding in the study of tumor formation, growth mechanisms, and drug responses. Three-dimensional tumor spheroids partially replicate the tumor microenvironment, allowing for better simulation of the interactions between tumor cells within tissues and multicellular tissue structures, providing more physiologically relevant models. Organoid models capture the essence of tumors at the organ level, offering a dynamic 3D model that can provide deeper insights into the mechanisms of tumor progression and has shown great potential in cancer drug screening. Tumor chips are a microfluidic technology that studies the physiological and pathological processes of organs by constructing microenvironments simulating human organs on microchips. Bioprinting technology allows for the precise 3D arrangement of cells and support structures, thereby simulating complex tumor tissues. There is a close connection between these three technologies. For example, 3D tumor spheroids can serve as the basic units for constructing tumor chips and organoids to simulate the tumor microenvironment and study tumor biological characteristics. At the same time, the development of organoid and tumor chip technologies also provides new tools and methods for the study of 3D tumor spheroids, such as precisely controlling culture conditions through microfluidic technology to enhance the physiological relevance and application range of 3D tumor spheroids.

Table 3 The Tumor Model Replicates the Tumor Microenvironment

In summary, driven by continuous advancements in science and technology and our understanding, the construction of tumor models has evolved from two-dimensional cell cultures to animal models, and now to organoids and other tumor models that can replicate the tumor microenvironment.

3D Tumor Spheroids

The 3D tumor spheroid is a model of three-dimensional cell culture utilized to mimic tumor tissues in vitro, created through the aggregation of cancer cells in a 3D cell culture environment. In contrast to conventional 2D cell culture models, this three-dimensional structure offers a more faithful representation of cell-cell interactions, cell-matrix interactions, and responses to external signals within tumor tissues. These spheroids can replicate cell-cell interactions,91 growth kinetics, and the deposition of new ECM similar to human solid tumors.92 Furthermore, accumulating evidence indicates that the dense ECM deposition and strong cell-cell contact within tumor spheroid aggregates can establish physical barriers resembling those present in the body, hindering the infiltration of drugs and immune cells, thus mirroring chemoresistance.93 By virtue of their tumor-like characteristics, tumor spheroids have become established models for drug screening, with the goal of reducing drug development expenses and unnecessary animal experimentation.94,95

Scaffold-Free 3D Tumor Spheroids

Scaffold-Free 3D tumor spheroids do not rely on external scaffold materials but instead naturally form spheroid structures through the inherent adhesion and proliferation capabilities of cells. These models typically utilize specially designed culture plates (such as low-adhesion or non-adhesive surfaces) or specific culture techniques (such as the hanging drop method, rotating culture) to facilitate spheroid formation.82 These methods generally promote cell-cell interactions by preventing cell-substrate interactions, thereby facilitating the rapid self-assembly of cells into multicellular aggregates. These approaches are relatively simple, offer high-throughput capabilities, and provide flexibility in combining various cell types.83 Scaffold-Free 3D tumor spheroids does not require complex scaffold materials; cells naturally aggregate to form spheroids, and the culture conditions are relatively simple with a high cell density. Scaffold-Free spheroids are typically cultured in standard 96- or 384-well plates, which also means they are more suitable for high-throughput drug screening. Since tumor cells are combined in the form of spheroids, when the size of these tumor spheroids exceeds 400 μm, they generate oxygen and nutrient gradients within the body, leading to core necrosis, a quiescent middle area, and a proliferative thick shell due to limited molecular transport. Roper et al96 stained tumor spheroids using immunohistochemistry and found that hypoxia-related biomarkers (CA9) marked distinct hypoxic and non-hypoxic areas (Figure 3D), showing rapid and stable growth in three-dimensional tumor spheroids (Figure 3E), which is crucial for studying the tumor microenvironment and drug response.

Figure 1 Schematic diagram of bioprinting application in tumor model construction.

Figure 2 Schematic diagram of bioprinting technology.

Figure 3 Manufacturing process of 3D tumor spheroids and related experimental results. (A) Culture protocol for 3D cell colony spheroids for implementation of 3D-ASM. Reproduced with permission from Lee SY, Hwang HJ, Lee DW. Optimization of 3D-aggregated spheroid model (3D-ASM) for selecting high efficacy drugs. Sci Rep. 2022;12(1):18937. http://creativecommons.org/licenses/by/4.0/.97 Copyright 2022, Springer Nature Ltd. (B) The surface of the 384 column plate and the bottom of the wet chamber. Reproduced with permission from Lee SY, Hwang HJ, Lee DW. Optimization of 3D-aggregated spheroid model (3D-ASM) for selecting high efficacy drugs. Sci Rep. 2022;12(1):18937. http://creativecommons.org/licenses/by/4.0/.97 Copyright 2022, Springer Nature Ltd. (C) Schematic diagram of combined proton beam-drug screening. Reproduced with permission from Lee DW, Kim JE, Lee GH, et al. High-Throughput 3D Tumor Spheroid Array Platform for Evaluating Sensitivity of Proton-Drug Combinations. Int J Mol Sci. 2022;23(2). http://creativecommons.org/licenses/by/4.0/.98 Copyright 2022, MDPI Ltd. (D) Immunohistochemical staining of hypoxic (CA9) markers of tumor spheroids at day 21 (scale:100µm). Reproduced with permission from Roper SJ, Coyle B. Establishing an In Vitro 3D Spheroid Model to Study Medulloblastoma Drug Response and Tumor Dissemination. Curr Protoc. 2022;2(1):e357.96 Copyright 2022, Wiley-Blackwell Ltd. (E) Representative images of 3D tumor spheroids over a 21-day period (scale:100µm). Reproduced with permission from Roper SJ, Coyle B. Establishing an In Vitro 3D Spheroid Model to Study Medulloblastoma Drug Response and Tumor Dissemination. Curr Protoc. 2022;2(1):e357.96 Copyright 2022, Wiley-Blackwell Ltd.

In recent years, researchers led by Li have developed a new method using hydrophobic silica nanoparticles to encapsulate liquid culture tumor spheroids and drug screening.99 Compared to traditional hanging drop methods, this approach not only enables real-time in situ imaging and motion control but also more closely replicates the microenvironment of in vivo tumor tissues. However, despite some achievements in tumor spheroid culture with these traditional methods, they face challenges in constructing scalable tumor spheroid models. These methods require a large number of cells and compounds as initial materials, which poses a barrier to applications involving rare cell-derived spheroids (eg, patient tumor biopsy samples). To overcome these limitations, researchers are exploring new strategies. One approach is to reduce the consumption of materials and cells to study spheroids from rare cells. Additionally, new methods such as using novel supporting matrices, microfluidic chip technologies are being explored to construct and control tumor spheroid models, better simulating in vivo physiological environments. In summary, the work of Li et al provides new insights and pathways for tumor spheroid culture, but further development and refinement are needed to better apply to the screening of rare cell-derived spheroids and personalized medicine. Researchers are committed to addressing these challenges to further advance tumor spheroid research.

In the study by Popova et al,100 a platform called Droplet Microarray (DMA) was used to form highly uniform, separated, and stable nanoliter droplets on hydrophilic spots in an ultrahydrophobic background, with each droplet capturing cells to serve as a reservoir for individual biological experiments. The mentioned DMA platform significantly reduces the use of compounds, reagents, and cells, thereby reducing screening costs while increasing throughput. Several common cancer cell lines were used to construct tumor spheroids (such as MCF-7, HEK293, and HeLa cells), with tumor spheroids formed in 100 nL droplets starting from only 150 cells per droplet within 24 to 48 hours. In the study by Xia et al,101 an ultrahydrophilic-ultrahydrophobic microarray platform combined with acoustic droplet technology was used. The acoustic droplet ejection device used could precisely inject drug solutions into the droplets, with the entire process completed within 20 milliseconds, achieving precise drug delivery to cell spheroids and improving the accuracy and efficiency of drug screening.

Scaffold-Based 3D Tumor Spheroids

Scaffold-Based 3D tumor spheroids are created by utilizing natural or synthetic biomaterials, such as extracellular matrix components and hydrogels, to support cell growth and tissue formation. Tumor cells grow on a matrix that mimics the tumor cell extracellular matrix. This scaffold not only promotes cell growth but also provides essential cues from the natural biophysics and biochemical environment, which are crucial for regulating tumor cell functions.84 Currently, scaffolds for in vitro cancer modeling have evolved from natural hydrogels (such as proteins, polysaccharides), synthetic hydrogels (such as polyethylene glycol, peptides), and hybrid materials (with high water content, tissue-like bioresponsiveness, viscoelasticity, and mechanical properties). Advances in microfabrication techniques in recent years have enabled more complex and precise hydrogel platforms, which can replicate the fundamental characteristics of the tumor microenvironment (TME). Compared to Scaffold-Free 3D tumor spheroids, these platforms offer better biocompatibility and structural stability due to the addition of scaffold materials, allowing for a more accurate simulation of the in vivo environment and the study of cell interactions.

Alginate, hyaluronic acid, and chitosan have been widely used to fabricate 3D tumor spheroids. For example, Sabhachandani and colleagues102 used a microfluidic-based droplet platform to generate single and co-culture spheroids of breast cancer in alginate hydrogel and explored their potential for therapeutic effect screening, showing that co-culture spheroids exhibited higher drug resistance than single culture spheroids. To better mimic the natural TME, further research by the team incorporated oxygen gradient generators and immune cells into their spheroid system to study the roles of hypoxia and immune cells in spheroid growth and drug response.103,104 A recent study by the team also explored the impact of immune cells on tumor response and progression.105

Matrigel is a commercially available basement membrane extract and has been the most commonly used three-dimensional cell culture matrix for constructing tumor spheroids in recent years. Matrigel’s origins date back more than 40 years and are primarily composed of mouse tumor cell laminin (laminin), type IV collagen, nidogen (entactin), and heparan sulfate proteoglycan extracted from Engelbreth-Holm-Swarm (EHS) tumors. It is a complex mixture of proteins that can solidify into a gel-like state at 37°C, simulating the natural environment of the extracellular matrix (ECM).106,107 Lee et al108 created large uniform dense spheroids by attaching Matrigel to a 96-well plate to simulate the tumor microenvironment of hepatocellular carcinoma. Through high-throughput drug screening, the tumor model was compared with traditional 2D tumor models, showing higher drug resistance, indicating that the newly constructed 3D tumor spheroid model is closer to in vivo conditions. Lee et al97 also used Matrigel to create an optimized 3D tumor spheroid model for drug screening. Cells and matrix gel mixtures were distributed on the surface of a 384-well plate using a 3D cell injector. After placing the 384-well plate in a humid chamber and refrigerating for 2 hours, the cells would aggregate at one location under CO2 incubator conditions. If the humid chamber and 384-well plate do not adhere properly, excessive gelation may occur. Under humid incubator and humid chamber conditions, the 3D cell mixture stabilizes after 20 minutes of gelation (5 minutes for gelation, 30 minutes for excessive gelation) (Figure 3A). In the case of a stable three-dimensional cell mixture, an aggregated cell colony can be observed on the surface of the 384-well plate, with no detached gel or free-floating cells observed at the bottom of the humid chamber (Figure 3B). Another research group98 used Matrigel to construct 3D tumor spheroids on a 384-well plate to study the therapeutic effect of proton beam therapy on squamous cell carcinoma of the head and neck and observed the therapeutic effect on the spheroids by controlling the irradiation dose (Figure 3C).

Currently, cancer cells are embedded in protein-based hydrogels (such as Matrigel,109 fibronectin,110 and collagen111,112 to prepare various tumor spheroids. These hydrogels combine specific cell adhesion sequences and enzyme-mediated matrix degradation, which are beneficial for cell proliferation. Aisenbrey et al113 discussed the limitations of Matrigel as a cell culture matrix and the applications and advantages of synthetic materials as alternatives to it in cell culture, regenerative medicine, and organoid assembly. Matrigel is a basement membrane matrix extracted from a specific type of mouse tumor, and its complex, ill-defined, and variable composition leads to uncertainty and lack of reproducibility in cell culture experiments. In the study by Kim et al,114 an extracellular matrix (ECM) hydrogel derived from gastrointestinal (GI) tissue was used as an alternative to traditional Matrigel. The results demonstrated t