Organ-on-a-Chip Systems for Modeling Tumor and Normal Tissue Microenvironments in Radiotherapy Research

Abstract

Radiation therapy precisely targets tumors with ionizing radiation, aiming to achieve local control while minimizing collateral damage to surrounding healthy tissues. Radiation research is often carried out in animal models, but these suffer from ethical issues, high cost of operation, low throughput, and low correlation to responses in humans. The advent of microfluidic organ-on-a-chip (OoC) technology offers a promising alternative to precisely and reproducibly model the physiology of different tissues in a laboratory setting. Furthermore, OoC models can be constructed from patient-specific tissues to tailor therapies while enabling fine control over relevant microenvironmental factors. This review covers emerging research at the intersection of radiation biology and microphysiological models, with a focus on the unique capabilities enabled by these advanced technologies.

Why develop advanced in vitro models for radiation therapy research?

Despite advances in detection and treatment, cancer remains the second leading cause of death worldwide [1]. Radiation therapy (RT) is used in about 50% of cancer cases, either with curative or palliative intent. Since its introduction in the late 1800s, RT has remained one of the most cost-effective and precise cancer treatments [2]. It can be delivered externally via devices such as linear accelerators (see Glossary) or internally through radiopharmaceuticals. While major progress has been made in physically targeting tumors while sparing healthy tissue, fewer advances have occurred towards developing compounds that synergize with radiation. This slow progress is largely due to the complexity of the physical, chemical, and biological responses that occur in living tissues following exposure to ionizing radiation.

Radiation biology examines how living systems respond to ionizing radiation at molecular, cellular, and tissue levels [3]. It encompasses DNA damage and repair, chromosomal aberrations, and cell death pathways, as well as how the tumor microenvironment (TME) and cellular radiosensitivity influence treatment outcomes [4]. Spanning from molecules to organisms, this field is essential for understanding radiation effects and guiding clinical strategies [5].

This review highlights how organ-on-a-chip (OoC) technologies can be harnessed to advance RT research by enabling precise control over tumor and normal tissue microenvironments. Compared to the previous reviews in the field [6][7][8], we provide a new perspective on this topic by focusing more specifically on how specific microenvironments can be engineered to modulate response to radiation (for instance, by controlling oxygen distribution and hypoxia, extracellular matrix composition, immune cell infiltration, and vascular dynamics). We also offer unique insights on current challenges and future directions for advancing the integration of radiobiology research with on-chip microphysiological systems.

Preclinical research in RT primarily relies on two-dimensional (2D) cell cultures and animal models. Although these models are commonly used, they suffer from critical shortcomings. For instance, 2D cell culture systems have low predictive power for identifying new radiomodulating agents through drug screens due to their lack of physiological relevance. Similarly, while 3D models such as organoids, spheroids, and other 3D in vitro models better mimic in vivo complexity, they still lack control over microenvironmental cues essential for predicting RT response [9][10][11]. Additionally, animal models are challenging to use because of ethical issues, high costs, and the poor translation of treatment outcomes from animals to humans due to species difference. Advances in microfabrication, microfluidics, biomaterials, and tissue engineering have enabled the development of physiologically relevant tissue models for preclinical research [12][13]. Among these, OoC systems—microfluidic platforms containing chambers for 3D human cell culture—offer dynamic, tunable environments that support nutrient perfusion, waste removal, and physiologically relevant architecture [14][15].

OoC technologies have been applied to model a variety of biological systems, including tumors-on-a-chip (ToC) and multi-organ-on-a-chip platforms that enable inter-organ communication. These models support animal-free, cost-effective, and high-throughput testing. Recent regulatory shifts, such as the FDA Modernization Act 2 and NIH initiatives supporting human-based alternatives to animal models, further promote the adoption of OoC systems [16].

OoC models are particularly valuable in radiobiology, where they can be used to investigate tumor responses to RT and radiation-induced injuries in normal tissues. A summary of techniques used for OoC modeling of radiation therapy is provided in Box 1. Incorporating patient-derived cells or organoids further enhances the relevance of these platforms for personalized RT studies. While primarily used in research, OoCs offer a new avenue for more accurate prediction of treatment responses and toxicities. Additionally, RT encompasses a range of distinct radiation types and delivery techniques, which are summarized in Table 1 and Box 2.

Box 1. Microfluidics and Organ-on-a-Chip Platforms for Radiotherapy Research.

Microfluidics offers powerful tools for engineering organ-on-a-chip (OoC) and tumor-on-a-chip (ToC) systems aimed at advancing preclinical radiotherapy (RT) research. These technologies enable the precise manipulation of microliter-scale volumes in miniaturized channels, facilitating cost-effective, high-throughput assays that conserve valuable patient-derived samples. The inherently laminar flow within microfluidic devices provides controlled and physiologically relevant shear stress, supporting the maintenance of complex tissue structures under dynamic culture conditions [86] [87]. Microfluidic platforms can be fully automated and are compatible with multiplexed analysis, offering improved reproducibility, faster processing times, and reduced contamination risk. Their compact and self-contained design also allows seamless integration with clinical or benchtop irradiation sources, enabling localized or systemic exposure models using various types of radiation [88]. Devices are typically fabricated using techniques such as photolithography, soft lithography, micromachining, or 3D printing. Polydimethylsiloxane (PDMS)—one of the most widely used materials in chip fabrication—is highly resistant to ionizing radiation and remains stable under therapeutic RT doses. Radiation penetration through microfluidic materials depends on the radiation type: while alpha particles and low-energy beta particles have limited penetration depth, photons, high-energy electrons, and protons can reach deep within tissues. For localized RT studies, radionuclides can be perfused directly into microchannels to deliver targeted alpha or beta particle exposure.

Table 1.

Overview of different RT modalities

Parameter	Radionuclide	X-ray tube	LINAC	Cyclotron	Synchrotron
Radiation Type	Alpha, beta, gamma	X-ray (photon)	Photon, electron	Proton, ion	Photon
Dose Range	0 – ~100 Gy (typical)	0 – ~100 Gy	0 – 100 Gy	0 – 100 Gy	0 – 100 Gy
Beam energy	10–1000 keV	50–400 keV	1–20 MeV	10–250 MeV	1–20 keV
Dose Rate (Gy/min)	0.1 – 10 Gy/min (typical)	0.1 – 10 Gy/min	1 – 10,000 Gy/min (e.g., FLASH)	1 – 1000 Gy/min	0.1 – 100,000 Gy/min
Field Size	0.1 – 100 cm2	0.1 – 100 cm2	0.1 – 100 cm2	0.1 – 100 cm2	0.1 – 100 cm2
Beam Type	Usually continuous (unpulsed)	Continuous	Pulsed	Pulsed or continuous	Pulsed

Box 2. Common RT Modalities and Their Compatibility with Organ-on-a-Chip Systems.

RT modalities include pre-clinical cabinet irradiators (e.g., X-rays, Cs-137 sources), clinical techniques such as intensity-modulated radiation therapy (IMRT), stereotactic body radiation therapy (SBRT), proton and carbon-ion therapy [89], and internal modalities such as brachytherapy [30] and radiopharmaceutical therapy [90]. Experimental modalities, including neutron and high-LET heavy ion irradiation (e.g., ⁵⁶Fe particles), have also been explored. Emerging technologies such as FLASH radiotherapy (FLASH-RT) offer ultra-high dose rates with promising therapeutic advantages [91][92]. Microfluidic systems are compatible with many of these sources due to their small footprint and design flexibility, enabling clinically relevant radiation exposures in vitro [93][8].

RT is highly effective in eradicating tumors but outcomes are strongly determined by the properties of the TME. OoC platforms enable precise control over factors such as oxygen distribution, extracellular matrix (ECM) composition, vascularization, and immune-stromal interactions, making them ideal tools for dissecting radiotherapy responses in a human-relevant, in vitro setting.

Despite being in the early stage of its development, microfluidic OoC/ToC technology shows potential in advancing radiobiological research by introducing innovative platforms for elucidating cellular, tissue, and organism-level responses to irradiation. The integration of OoC/ToC platforms with radiation treatment has been explored by different research groups, using a variety of approaches and by leveraging the potential of these technologies to engineer key microenvironmental factors for preclinical radiobiology research. Figure 1 gives an overview of the integration of ToC and OoC models in radiation biology.

Figure 1. Radiotherapy using Organ-on-a-Chip Models.

Schematic overview depicting the application of radiation therapy to tumor-on-a-chip and normal tissue-on-a-chip models. These platforms can culture patient-derived organoids, cell lines, primary cells, or stem cells in precisely engineered 3D microenvironments to replicate relevant physiology. Tumor-on-a-chip systems model key features of the tumor microenvironment, while tissue chips enable the study of radiation-induced toxicity in healthy organs. Perfusion capabilities support dynamic culture conditions. Post-radiation assessments include a range of cellular, molecular, and functional analyses. This figure illustrates the potential of organ-on-a-chip technologies to provide human-relevant platforms for radiotherapy research.

Tumor-on-a-chip systems as radiobiological models

A tumor-on-a-chip (ToC) is an advanced in vitro preclinical model designed to recapitulate the most salient physiological features of the TME within a microfluidic device. This technology offers significant advantages for studying how tumors respond to radiation and improving the efficiency and effectiveness of cancer treatments [17]. ToC models address the shortcoming of current preclinical models by providing more controllable and in vivo-like culture microenvironment for cells and tissues while also enabling the use of patient-derived cells to individualize patient treatments [18]. Moreover, compared to animal models, a ToC is a high-throughput platform capable of evaluating many combinations of radiotherapy and chemotherapy (CT) [19]. Additionally, advanced ToC models can incorporate different cell types, including cancer cells, endothelial cells, stromal cells, immune cells, and fibroblasts, in a spatially and temporally controlled manner. Three major applications of ToC models are for (i) studying cancer progression and resistance, (ii) screening anti-cancer treatments in vitro, and (iii) creating patient-specific tumors for personalized therapy. In this section, we focus on the role that ToC platforms can play in modeling radiotherapy and discovering agents that can modulate its effects. Additionally, we discuss the influence of tumor-intrinsic factors, such as oxygen, pH, and immune infiltration.

Several studies have analyzed the effects of RT using microfluidic tumor models. Staicu and colleagues [20] developed a miniaturized lab-on-a-chip device to create a 3D model that mimics the in vivo environment for testing the effects of various X-ray radiation doses on tumor cells, specifically human melanoma cells. This setup maintains optimal temperature, oxygen, and pH conditions during observations. Additionally, the glass device incorporates chromatographic sensors for simultaneous assessment of the delivered dose and the biological impacts of ionizing radiation on cancer cells. Exposure of these devices to X-ray induced Ag ion photoreduction, causing a color change for direct radiation dose verification. This glass biochip was used as a proof of concept to study the response of human melanoma cells to various X-ray doses according to their production of reactive oxygen species (ROS).

RT is crucial for treating head-and-neck squamous cell carcinoma (HNSCC), but radioresistance remains a major hurdle. Carr and coworkers [21] used a microfluidic platform to maintain HNSCC biopsies from 35 patients and assess their response to 2–40 Gray (Gy). Release of lactate dehydrogenase (LDH, a cell death marker) increased significantly at 40 Gy, while immunostaining showed a dose-dependent rise in apoptosis. In a follow-up study [22], biopsies from five patients were exposed to 0–20 Gy in a microfluidic chip. Higher doses resulted in increased apoptosis, DNA damage and decreased proliferation (Figure 2A). Compared to their previous study, this study highlighted inter-patient variability, supporting the potential of microfluidic models for personalized RT using multiplex biomarkers.

Figure 2. Integration of radiation therapy with tumor-on-a-chip models.

A) Head-and-neck cancer-on-a-chip. (a) Image depicting the assembly of a microfluidic glass chip connected to inlet and outlet tubing. (b) The microfluidic chip containing HNSCC tissue is positioned within a custom-built Perspex holder and irradiated using a linear accelerator with dual lateral opposing 8×8 cm fields at a dose rate of 600 MU/min. Reproduced with permission from [22]. B) Soft tissue sarcoma-on-a-chip. (a) Dose modulation strategy using an orthovoltage beam and a 3D-printed holder with a narrow slit to position a 3 mm-thick lead plate, enabling precise and localized dose delivery to microfluidic cultures. (b) CAD rendering of the microfluidic platform and a representative image of the chamber seeded with STS93 spheroids. Reproduced with permission from [27]. C) Pancreatic cancer-on-a-chip. (a) Schematic of a transwell-style microfluidic platform incorporating human fibroblasts (MRC-5) and human pancreatic adenocarcinoma (Panc-1) spheroids. (b) Fluorescence microscopy images of Panc-1 spheroids stained with Calcein-AM (live cells, green) and propidium iodide (dead cells, red) at 72 h post-treatment under varying radiation exposure conditions. (c) Confocal images showing DNA damage (γH2AX, magenta) and nuclei (SYTOX Green) in Panc-1 spheroids of different sizes, either untreated or exposed to neutron radiation alone or combined with BNCT, 24 h post-irradiation. Reproduced with permission from [28]. D) 3D bioprinted glioblastoma-on-a-chip. (a) Cross-sectional schematic of native glioblastoma (GBM) tissue. (b) Illustration of the GBM chip fabrication method using distinct bioinks and biomaterials to create a compartmentalized architecture. (c) Prototype GBM chip images showing BdECM bioink with HUVECs (magenta) or GBM cells (blue), viewed from above (top) and angled (bottom). Scale bar: 2 cm. Reproduced with permission from [29].

Combining RT with chemotherapy (CT) is a common treatment strategy [23][24] that can be studied using physiologically relevant ToC systems [25]. Patra and colleagues [26] used a PDMS microfluidic device to culture uniform soft-tissue sarcoma (STS) spheroids and treat these with RT and doxorubicin. In one model, CT induced cell death by apoptosis whereas RT triggered alternative death pathways. Another p53-mutant model was resistant to both treatments. This approach supports the use of ToC systems to screen RT-drug combinations, especially for rare cancers. In a follow-up study [27], they used a high-throughput chip to grow STS spheroids and screen radiosensitizing and radioprotecting agents. A movable lead shield was developed to spatially modulate RT across the chip. Talazoparib showed synergy with RT, while Pazopanib and AZD7762 had additive effects (Figure 2B), demonstrating the potential of ToC models for optimizing RT-drug regimens.

Boron Neutron Capture Therapy (BNCT) combines boronated compounds and neutron irradiation to selectively target cancer cells. Yu and colleagues [28] developed a ToC model to study BNCT’s effectiveness in pancreatic cancer using a high-throughput 3D spheroid platform. Real-time monitoring revealed that larger, hypoxic spheroids were more resistant, with HIF1-α signaling associated with reduced sensitivity. Fibroblast infiltration and β-catenin translocation also contributed to resistance, highlighting the roles of hypoxia and fibrosis in mediating BNCT efficacy (Figure 2C).

Yi and colleagues [29] developed a bioprinted glioblastoma-on-a-chip model to create patient-specific models. The model reconstituted a glioblastoma tumor by mixing patient-derived cancer cells, vascular endothelial cells, and brain extracellular matrix and was organized in a cancer–stroma concentric-ring structure with a radial oxygen gradient. The model replicated the properties of native glioblastomas and successfully reproduced patient-specific resistance to chemoradiation with temozolomide. It was also used to identify effective drug combinations, offering a potential method for screening second-line treatments for glioblastoma patients resistant to standard therapies (Figure 2D).

Chermat and colleagues [30] developed a microfluidic device to test various brachytherapy (BT) schemes using an in vitro model. This PDMS-based BT-on-chip system allows for precise insertion of radioactive iodine-125 BT seeds, accurate dose calculations using TG-43 formalism, and short-term culture of hypoxic spheroids. The platform was used to study the influence of dose rate and hypoxia on tumor response. Hypoxia in the spheroids was confirmed, and treatment responses were assessed through DNA damage and cell survival assays, revealing dose-dependent effects [31].

In summary, these radiobiological studies demonstrate the remarkable potential and versatility of ToC platforms in advancing radiation research. Table 2 lists key aspects of these studies, along with several other notable works.

Table 2.

Selected radiobiological studies utilizing ToC/OoC models

Tumor-on-a-chip as a model for RT
Cancer model	Aim of study	Cells	Radiation type and dosage	Drug combination	Assessment techniques	Ref.
Melanoma	Simultaneous dosimetry measurements and evaluation of radiation effects on cancer cells.	A375 melanoma cells	280 kVp X-ray 0–8 Gy	No	ROS, Viability and proliferation test by MTS assay	[20] 7/4/2025 6:37:00 AM
Head and neck cancer	A patient tumor-on-a-chip system for personalized RT	Head and neck squamous cell carcinomas (HNSCC)	120 kVp X-ray 5 × 2 Gy	No	Proliferation (Ki67 and BrdU), DNA damage response (γH2AX) and caspase-dependent apoptosis	[94]
Soft tissue sarcoma (STS) spheroids	Finding potential drugs used in combination with RT	Tissue sarcoma tumor spheroids	140 kVp X-ray 0.5 – 8 Gy	Talazoparib, Pazopanib, AZD7762	Comet assays. clonogenic assay	[27]
Pancreatic tumor spheroids-on-a-chip	High-throughput 3D tumor spheroid platform for BNCT	Pancreatic tumor	Neutron (dosimetry not specified)	No	Clonogenic assay, proliferation marker (Ki67)	[28]
Human head and neck squamous cell carcinoma-on-a-chip	Measuring the response of HNSCC to irradiation	HNSCC biopsies from five patients	6 MV X-ray linac Single-dose irradiation (0, 5, 10, 15, and 20 Gy)	No	LDH release, IHC of CK, cleaved-CK18, γH2AX, and Ki-67. DNA fragmentation via TUNEL	[22]
Soft-tissue sarcoma spheroids-on-a-chip	Chemoradiotherapy testing-on-chip	Soft-tissue sarcoma spheroids	6 MV X-ray linac Single doses of 0.5,2, and 8 Gy	Yes. doxorubicin at 2 μM and 20 μM.	Flow cytometry and clonogenic assay	[26]
Tumor spheroid-on-a-chip	Brachytherapy on-a-chip	Spheroids from SK-LMS-1 (sarcoma), FaDu (head and neck), HCT116 (colon)	I-125 seeds (28 keV gamma)	No	Comet assay, clonogenic assay, immunofluorescence to measure γH2AX	[30]
Organs-on-a-chip to model radiotoxicity
Organ model	Aim of study	Cells	Radiation type and dosage	Drug combination	Assessment techniques	Ref.
Human Gut-on-a-Chip	Intestinal injury induced by acute exposure to γ-radiation	Human intestinal epithelium and endothelium	γ-radiation (Cs-137) 8 Gy	Prolylhydroxylas e inhibitor dimethyloxalylglycine (DMOG)	TUNEL staining, ROS, lipid peroxidation, 53BP1 + punctate staining	[33]
Human lung-on-a-chip	Studying radiation-induced lung injury (RILI)	Lung microvascular endothelial cells, alveolar epithelial cells	γ-radiation (Cs-137) 12 to 16 Gy	Lovastatin and prednisolone	Immunostaining for DNA damage, ROS detection, Cytokine analysis, Bulk RNA sequencing	[32]
Liver-on-a-chip	Study of drug conversion and radiation protection of liver tissue	Human hepatic carcinoma cells (HepG2) and human mammary epithelial (M10)	γ-radiation (Cs-137) dose rate of 86 cGy min-1 2 Gy	Amifostine	Live/Dead stain, Quantification of radiation damage by the micronuclei count	[95]
3D microvasculature-on-a-chip microfluidic =	Impact of radiation on microvasculature	Human umbilical vein endothelial cells	6 MV X-ray linac 0–25 Gy	No	cellular apoptosis, vessel tight adherens junction breakage, DNA double strand break, and repair	[35]
Oral mucositis on a chip	Modeling mucositis induced by chemo- and radiation treatments	Keratinocytes, fibroblasts and endothelial in 3D collagen gel	UV radiation (mimicking RT)	Cisplatin	Cytotoxicity, mucosal barrier function, cell damage and viability, LDH	[41]
A multi-niche microvascularized human bone marrow (hBM) on-a-chip	Modeling response to RT of human bone marrow using 96-well microfluidic chip	Human mesenchymal stromal cells, endothelial cells, CD34+ hematopoietic stem cells	X-ray 5 Gy	No	LDH, cytokines detection, apoptosis-induced DNA fragmentation via TUNEL assay	[40]
Human fibroblasts on a chip (skin model)	Use microfluidic chips to study the effects of ultraviolet lights on human fibroblasts	NIH/3T3 fibroblasts	UVB radiation doses ranging from 0 to 0.6 J/cm2	No	Morphology and fluorescence-based cell damage rate	[96]
BBB-on-a-chip	A high-throughput human OoC system to study the effect of simulated deep space radiation on BBB function	Human cerebral microvascular endothelial cells, peripheral blood microvascular endothelial cells, induced pluripotent stem cell-derived astrocytes	600 MeV/n 56Fe particles and simplified simulated galactic cosmic rays	No	Vascular permeability, oxidative stress, inflammation, pro- inflammatory cytokines	[42]
BBB-on-a-chip	To investigate the mechanisms of radiation-induced brain injury (RIBI) using a human BBB MPS, focusing on mitochondrial-mediated sterile inflammation	Human brain microvascular endothelial cells (HBMECs), astrocytes, pericytes	X-ray and γ-ray, acute exposure (10 Gy)	Abrocitinib (JAK1 inhibitor), Idebenone (mitochondrial protectant)	TEER measurement, immunofluorescence (ZO-1, GFAP), ROS assays, γ-H2AX staining, proliferation assays, mitochondrial membrane potential, qRT-PCR, Western blot, cytokine ELISA	[97]
Multi-organ-on-a-chip (bone marrow, liver, heart)	Developing an OoC model to study deep radiation effect on astronauts	Heart: cardiomyocytes (iPSC-derived) mixed with primary human cardiac fibroblasts, Bone marrow: iPSCs, HUVECs, and CD34+ cord blood-derived hematopoietic stem and progenitor cells. Liver: iPSC-derived hepatocytes and primary human dermal fibroblasts.	Delivering the same dose (5 Gy) of high-LET neutrons either acutely as a single dose or over 2 weeks (dose rate ≈0.04 Gy per day)	No	Flow cytometry, cytokine secretion, immunostaining, cardiac tissue contraction evaluation, tissue morphology	[43]
Vascularized Bone marrow-on-a-chip	Model human hematopoietic vascular niche and simulate radiation-induced injury using a multi organ model	CD34+ HSPCs, HUVECs, MSCs, fibroblasts	Pencil beam scanning proton radiation: 0.2, 2, and 20 Gy	Cisplatin (5, 15 μM); 5-FU	Immunofluorescence, RNA-seq, colony-forming unit (CFU) assay, cytokine profiling	[44]

Organ-on-a-chip systems as models of normal tissue radiotoxicity

Despite its potent anti-tumor action, radiation can also induce harmful effects in normal tissues. Exposure to low-dose radiation commonly occurs because of medical procedures, job-related tasks, or background radiation, and can increase the risk of cancer and germline inheritable mutations. Higher doses (>1 Gy) lead to acute, deterministic effects and are primarily a concern for patients undergoing RT due to unavoidable damage to surrounding healthy tissues. At the highest doses, radiation can be lethal by impairing the hematopoietic, gastrointestinal, or cerebrovascular systems. On a cellular level, radiation causes DNA damage, cell dysfunction, and tissue injury. OoC systems mimic the physiology of tissue microenvironments and can be used to study radiation-tissue interactions, dose-response dynamics, tissue-specific effects, and radioprotectors. Given existing risks, there is a need for more comprehensive data on both acute and stochastic effects of radiation on normal tissues. This section explores microfluidic tissue models designed to reveal the balance between therapeutic efficacy and normal tissue damage in radiotherapy or following accidental exposures.

Radiation-induced lung injury (RILI) is a serious complication of thoracic RT that encompasses both pneumonitis and fibrosis. A human lung alveolus-on-a-chip was developed to model acute RILI, simulating alveolar epithelium, pulmonary endothelium, and breathing motion. Within 6 hours of gamma radiation, the model exhibited DNA damage, cellular hypertrophy, inflammatory cytokine release, and barrier dysfunction, with endothelium more affected than epithelium [32]. The model’s dose sensitivity mirrored human lung responses and was used to test lovastatin and prednisolone for mitigating RILI effects (Figure 3A). The study showed that OoC models provide a human-relevant tool for studying acute RILI and evaluating potential countermeasures.

Figure 3. Organ-on-a-chip models of radiotoxicity.

A) Lung-on-a-chip model of acute radiation injury. (a) Schematic of the alveolar chip design with a confocal z-stack image highlighting endothelial tubule formation (scale bar: 100 μm). (b) Immunofluorescence staining for 53BP1 (green), indicating double-stranded DNA breaks two hours post-radiation. Nuclei are counterstained with DAPI (white). Scale bar: 20 μm. Reproduced with permission from [32]. B) Gut-on-a-chip model. (a) Schematic illustrating the initial seeding of human intestinal epithelial and endothelial cells on opposite sides of a matrix-coated porous membrane within a dual-channel microfluidic device. (b) Development of a villus-like epithelium in the upper channel interfacing with a planar endothelium forming a lumen in the lower channel. (c) Fabricated microfluidic setup for the gut-on-a-chip model. (d) Representative confocal immunofluorescence image showing a cross-sectional view of the chip, with villin (green) marking the apical brush border of the epithelium and VE-cadherin (red) labeling endothelial adherens junctions. Scale bar: 100 μm. Reproduced with permission from [33]. C) Microvasculature-on-a-chip model. (a) Human umbilical vein endothelial cells (HUVECs) are embedded in a fibrin gel within the central microfluidic channel (red), while normal human lung fibroblasts (NHLFs) are cultured in the adjacent side channels (blue). (b) The microvasculature-on-a-chip model at day 0 and day 5. Radiation is administered on day 5 following microvascular network formation to assess radiation-induced vascular damage. Reproduced with permission from [35]. D) Bone marrow-on-a-chip model. (a) Overview of the process to create a bone marrow-on-a-chip system, where engineered bone marrow (eBM) is first formed in vivo within a PDMS device and subsequently maintained in a microfluidic culture platform. (b) Eight weeks post-implantation, a white cylindrical bone structure forms with visible pink marrow inside the eBM. (c) Microfluidic chip used to culture the explanted eBM in vitro. Scale bars: 2 mm. Reproduced with permission from [38]. E) Blood–brain barrier-on-a-chip model. (a) Schematics for the BBB-on-a-chip model and its components. (b) Top: Irradiation details for the BBB model. Bottom: Confocal image showing PECAM⁺ endothelial cells (red) in the top channel and GFAP⁺ astrocytes (green) in the bottom channel. Reproduced with permission from [42]. F) Multi-organ-on-a-chip model. (a) Individual human liver, heart, and bone marrow tissues are developed and matured over 4–6 weeks, then integrated into a multi-organ-on-a-chip platform. (b) Subjection of the platform to either acute or prolonged neutron radiation exposure at a total dose of 0.5 Gy over a two-week period. Reproduced with permission from [43].

OoC models are a valuable in vitro tool for studying intestinal injury from high dose radiation and testing protective drugs. A human gut-on-a-chip microfluidic device was used to model radiation-induced cell death and test pharmaceutical countermeasures [33]. The model, lined with human intestinal epithelial and vascular endothelial cells, accurately replicated intestinal physiology. Exposure of the chip to gamma radiation increased reactive oxygen species, cytotoxicity, apoptosis, DNA fragmentation, and disrupted intestinal barrier integrity. Pre-treatment with the drug dimethyloxaloylglycine (DMOG) significantly reduced these effects (Figure 3B).

Ionizing radiation can impair microvasculature, leading to organ dysfunction. OoC models have been used to study and mitigate these vascular effects. Millet and colleagues [34] developed a human-derived microvascular microfluidic lumen model in which HMVEC-L cells were grown on microfluidic chips. They identified 35 potential biomarkers of ionizing radiation exposure using mass-spectrometry-based proteomics. This study demonstrates the utility of humanized OoC systems for biomarker discovery and testing countermeasures. In another study, 2D and 3D microvasculature-on-a-chip models were used to investigate vascular responses to ionizing radiation [35]. Differences in apoptosis and DNA damage highlighted the advantages of 3D systems for studying radiation effects (Figure 3C). Additionally, the effects of gamma irradiation on endothelial cells were investigated using 3D blood vessel models, revealing collapsed barriers and reduced angiogenic potential [36]. Another 3D microvasculature-on-a-chip study showed acute and subacute vascular damage, including reduced area, branch length, and function in both mature and new vessels [37]. Together, these studies highlight the value of OoC models in advancing understanding of radiation-induced vascular toxicity and supporting countermeasure development.

Bone marrow is highly sensitive to ionizing radiation, but traditional in vitro models lack the complexity of animal models. Bone marrow-on-a-chip models offer a more advanced platform to replicate its physiology and study radiation effects. For instance, Torisawa and others [38] developed a bone marrow–on–a–chip by implanting engineered bone in a mouse and later culturing it in a microfluidic device. The system maintained hematopoietic stem and progenitor cells for at least one week and replicated organ-level radiation toxicity, confirming the protective effects of countermeasure drugs, outperforming conventional cultures (Figure 3D). In a related study, a similar chip supported mouse hematopoietic cells for two weeks; gamma radiation reduced leukocyte output, while certain drugs restored hematopoiesis, demonstrating the device’s potential for assessing radiation effects and drug efficacy [39]. In a third study, a high-throughput human bone marrow-on-a-chip was developed to recreate endosteal, central marrow and perivascular niches, enabling detailed analysis of hematopoietic stem cell behavior and radiation responses [40]. In this model, reduced stem cell apoptosis was observed in endosteal niches, pointing to the protective effects of the osteoblast surface. These studies demonstrate how advanced bone marrow-on-a-chip systems can be used to study radiation effects and evaluate therapies in a human-relevant context.

Oral mucositis (OM) affects about 70% of head-and-neck cancer patients undergoing chemotherapy and/or radiation, with no widely effective prevention available. Ly and colleagues [41] developed the OM chip, a microfluidic oral mucosa triculture tissue construct that models OM induction and recovery from cancer treatments. The chip included a keratinocyte layer atop a fibroblast and endothelial cell-embedded collagen gel, stabilized by photocrosslinking to maintain tissue geometry and barrier function for over 18 days. Following exposure to cisplatin and radiation, the model displayed markers of tissue damage and recovery, providing a platform for studying OM mechanisms and potential treatments.

Cosmic radiation presents a major risk to the central nervous system (CNS) and cognitive functions during deep-space missions. A high-throughput human OoC system was used to simulate the effects of deep space radiation on blood-brain barrier (BBB) integrity and astrocyte behavior. Exposure to heavy ions such as ⁵⁶Fe increased vascular permeability, oxidative stress, and inflammation, and delayed astrocyte activation [42]. Initially exacerbating BBB disruption, astrocytes later reduced oxidative stress and inflammation, indicating a dual-phase response (Figure 3E). These results provide insight into potential strategies for mitigating CNS damage during spaceflight.

Multiple OoCs can be linked together to simulate organ-organ interactions, creating more physiologically relevant models [43][44]. Tavakol and colleagues [43] developed a multi-organ-on-a-chip platform mimicking human bone marrow, heart, and liver to study cosmic radiation. They found greater tissue damage and unique gene expression patterns following protracted neutron exposure, offering insights into prolonged radiation effects and potential space radioprotection (Figure 3F).

In summary, these studies highlight the potential of OoC models to enhance our understanding of radiation effects on normal tissues, predict healthy tissue responses, and guide the development of radioprotection strategies. Table 2 summarizes key findings from notable radiotoxicity studies.

Assessing cellular responses to radiation

Various assays have been developed to assess cellular responses to ionizing radiation, targeting key functions like proliferation, apoptosis, DNA damage, oxidative stress, migration, cytokine release, and gene expression. Integrating these assays into on-chip models is essential for evaluating radiation effects. It is important to distinguish between assays that can be performed directly on-chip and therefore preserve spatial context, such as staining and microscopy, and those that require cell harvesting for further analysis.

In situ assays include γ-H2AX immunostaining, which is used to measure double-strand break DNA damage; fluorogenic probes for reactive oxygen species, which measures oxidative stress; TUNEL staining, which measures radiation-induced apoptosis programmed cell death; and cell migration and invasion assays, which monitor metastatic potential. A few more assays can be performed non-invasively from chip effluent, such as detection of LDH for evaluating cell membrane integrity and viability, and cytokine release assays to measure inflammatory responses.

In contrast, other assays require the retrieval of the cells from the device. The clonogenic assay is the gold standard endpoint for evaluating the proliferative fate of cells after RT [45]. This assay requires cells to be harvested from the chip post-RT and seeded sparsely in 2D or 3D in standard culture plates. The Comet assay is another method for measuring DNA fragmentation in cells based on gel electrophoresis. RNA sequencing is used to measure bulk gene expression in the chip. Finally, cell cycle analysis is typically performed by flow cytometry of the irradiated cells.

Microfluidic-based sensors have also been developed for radiation studies [37]. Examples include microfluidic chip calorimeters for precise dosimetry [46], fixed cytometer chips for real-time single-cell apoptosis monitoring [47], and miniature fluorescence cytometers for quantitative radiation dose evaluation via γ-H2AX assays [47]. Additionally, microfluidic radiobioassays are powerful tools for studying molecular uptake and kinetics in cells, aiding the development of novel targeted radiopharmaceuticals [48][49]. These studies highlight the potential of microfluidic devices in advancing radiation dosimetry and cellular response analyses.

In summary, the assays discussed above highlight the complementary roles of on-chip and off-chip approaches in evaluating radiation responses within OoC systems. Clonogenic survival and DNA damage assays are particularly suited for slow-turnover cells, while apoptosis, LDH release, and viability assays are more relevant for assessing acute normal tissue radiotoxicity in rapidly cycling cells. On-chip tools such as integrated ROS sensors and live-cell imaging provide real-time insights under controlled microenvironments, whereas off-chip analyses like flow cytometry, RNA-seq, and proteomics enable deeper mechanistic investigations.

Engineered microenvironment effects as radiation response modifiers

A reciprocal relationship exists between the TME and RT: the TME influences efficacy of RT, while RT, in turn, remodels the TME by altering the vasculature and inducing inflammation [50]. Key factors such as ECM composition, hypoxia, vascular structure, and immune infiltration play critical roles in radioresistance. ToC platforms allow precise control and investigation of these parameters, enabling deeper insight into how the TME modulates RT outcomes [51][52]. Figure 4 illustrates the impact of various TME parameters on RT outcomes and highlights the role of microfluidic models in controlling these parameters. Here we discuss the role of some key microenvironment factors.

Figure 4. Radiation therapy and tumor microenvironment effects.

Characteristics of the tumor microenvironment and its impact on radiation therapy response. Factors such as the extracellular matrix, immune infiltration, hypoxia, oxygen gradients, and vasculature microenvironmental contribute substantially to the outcome of radiotherapy.

Oxygen, Metabolites, Nutrients, and Waste Products

Tumor hypoxia significantly affects the effectiveness of treatments like radiotherapy, chemotherapy, and surgery [53]. The impact of ionizing radiation on tumor cells is crucially dependent on oxygen levels in the TME. Under normoxic conditions, where oxygen is adequate, DNA lesions that result from the direct and indirect action of ionizing radiation react with oxygen to form more complex peroxyl lesions that cannot be easily repaired—the damage is said to be “fixed” by oxygen [54]. In contrast, under hypoxic conditions, DNA radicals can be chemically restored by cellular glutathione (GSH) without interference from oxygen, leading to increased tumor cell survival [55]. Hypoxia also alters gene expression through HIF transcription factors, further contributing to radioresistance. Due to these effects, oxygen is one of the most potent modulators of radiation sensitivity, capable of boosting the effective delivered dose up to threefold.

Understanding the interaction between hypoxia, radiation, and cellular responses is crucial for developing therapies that target hypoxic tumor cells and enhance RT efficacy. Microfluidic ToC models can create a hypoxic tumor microenvironment, using strategies like oxygen scavengers [56], impermeable materials, or N₂-supplemented media to establish oxygen gradients [57]. These systems mimic in vivo heterogeneity and support investigation of RT and other cancer therapies.

Tunable oxygen conditions have been incorporated into several microfluidic models to investigate the role of hypoxia in RT. In one approach, a device was developed to culture, treat, and analyze up to 240 naturally hypoxic “jumbo” sarcoma spheroids [58]. These spheroids expressed typical hypoxia markers such as carbonic anhydrase IX (CAIX). Treatment of these spheroids with combined RT and the hypoxia-specific prodrug tirapazamine (TPZ) selectively targeted hypoxic cells, demonstrating the utility of the ToC platform for hypoxia studies. In another system, high-throughput on-chip gas mixing was used to generate spatial oxygen gradients for breast tumor spheroids [59]. These gradients influenced ROS generation and the cytotoxic effects of doxorubicin and TPZ.

In addition to oxygen, the effectiveness of RT is influenced by the local concentration of metabolites, nutrients, and waste products [60]. Metabolites, such as ROS, can enhance radiation-induced DNA damage, while hypoxia-related metabolites may lead to radioresistance by stabilizing the hypoxic environment. The availability of nutrients such as glucose and amino acids influences the ability of cells to repair radiation damage. Conversely, waste products such as lactic acid create an acidic microenvironment that may reduce radiation efficacy by altering cellular metabolism [61][62]. ToC models offer a promising platform for studying these effects, particularly by allowing the control and manipulation of gradients of these parameters within a microfluidic environment.

Extracellular matrix (ECM)

The ECM of solid tumors is composed of collagens, fibronectin, elastin, and laminins, which are produced by cancer cells and cancer-associated fibroblasts (CAFs). The properties of the ECM vary across cancer types, acting as a barrier to oxygen and nutrients while promoting epithelial-to-mesenchymal transition (EMT) [51][63]. It also modulates the TME through signaling and increases in stiffness during tumorigenesis, promoting malignancy. Understanding ECM dysregulation may reveal new therapeutic targets [64].

The ECM also significantly influences RT outcomes. While it does not block radiation directly, its composition and structure are altered by exposure, affecting tumor cell behavior. An in vitro hydrogel model mimicking normal tissue showed that radiation-induced ECM changes promoted tumor cell proliferation, actin alignment, and invasiveness [65]. Incorporating ECM into OoC platforms could enhance the modeling of radiation-induced remodeling and its impact on tumor progression.

Different hydrogel-based ECMs such as fibrin, gelatin methacryloyl, collagen, and Matrigel are used to model 3D tissue inside microfluidics chips [66][19]. This approach considers cell-ECM and cell-cell interactions within a 3D microenvironment, offering a more physiologically relevant model for studying various biological processes and disease mechanisms [67][68]. ToC models can precisely control the alignment of the ECM fibers, providing a platform to systematically investigate its effects on cancer cell invasion and migration [69].

Microfluidic systems can also be imaged with high resolution and in real-time to monitor cellular responses to RT in the presence of a well-defined ECM. This can help in understanding how different ECM compositions and structures affect radiation sensitivity, resistance mechanisms, and migration behavior.

Vasculature

Tumor vasculature is vital in contributing to tumor progression and radiotherapy response. Not only is the vasculature function to deliver oxygen and nutrients and remove metabolic wastes, but it is also a key route for tumor metastasis, a process that can be modeled using microfluidics devices [69]. Metastasis of cancer via the bloodstream accounts for around 90% of cancer-related deaths. Thus, studying the development, flow, and intravasation/extravasation of cells from the vasculature is crucial to analyzing tumor progression and developing treatment options for cancer patients. Microfluidic-based models can be used to recreate the in vivo-like microenvironment to study the effect of RT on the vasculature and the ability of tumor cells to metastasize. Several studies have indicated that RT promotes tumor cell migration and metastasis [70][71].

Hypofractionated radiotherapy using large doses of radiation delivered in a few fractions has recently emerged as a viable alternative to conventional fractionation. Clinical evidence demonstrating excellent tumor control for hypofractionated RT for various cancers has led to speculation about the biological mechanisms involved as its mode of action. Among different factors, the effect of large-dose fractions on the tumor vasculature has been evoked a potential mediator of tumor control [72]. The ability of ToC to precisely control the vascularization of tumors [73] makes it the ideal platform to investigate these questions using this RT module.

Infiltration of immune cells

RT has both immunostimulatory and immunosuppressive effects, making its interaction with the immune system critical to treatment outcomes [74]. RT can enhance tumor killing by promoting antigen presentation, activating dendritic and cytotoxic T cells, inducing the stimulator of interferon genes (STING) pathway, and increasing immune cell infiltration through modulation of vascular structures, cytokine secretion, and adhesion molecule expression [75]. The role of the patient’s immune system in RT responses is also noteworthy due to studies which have demonstrated abscopal effects. This occurs when a systemic effect is achieved by combining local RT and systemic immunotherapy [76].

On the other hand, irradiation can result in immune suppression through activation of cells such as T regulatory cells, tumor-associated macrophages, and myeloid derived suppressor cells, and by killing tumor-infiltrating lymphocytes. Therefore, it is vital to understand the factors leading to immunological activation and suppression [77].

Microfluidic ToC platforms have been used to model these dynamics. For instance, neutrophil behavior was studied using a tumor-immune microenvironment chip, revealing extracellular trap formation that facilitated tumor invasion [78]. Triple-negative breast cancer (TNBC) is a clinically relevant subtype with a distinct TME that affects radiation response, making it well-suited for OoC-based modeling. While TNBC patients often receive chemotherapy, surgery, and RT, recurrence within five years remains common. Preclinical studies suggest that immune cell infiltration at irradiated sites may promote tumor cell recruitment, though mechanisms are unclear, highlighting the need for controlled, physiologically relevant models [79].

As RT reshapes the TME from “cold” to “hot,” the addition of immune checkpoint inhibitors holds strong therapeutic potential. However, the complexity of RT–immune interactions requires advanced models. OoC systems incorporating immune components, such as tumor-infiltrating lymphocytes or myeloid cells, provide valuable mechanistic insights and enable preclinical testing of combination therapies [76]. Clinically, RT may enhance immunotherapy by increasing antigen presentation and promoting inflammation. Recent studies suggest that sequential RT followed by immunotherapy may be more effective than concurrent administration [80,81].

Concluding remarks and future perspectives

OoC models play a pivotal role in recreating tissue-level functionality for RT research, including assessing tumor responses and normal tissue complications. The versatility of microfluidic chips with respect to different types of radiation highlights their potential in radiobiology research. However, compared to other models, relatively little radiobiology research has been conducted using OoC and ToC models, highlighting the need for further exploration and investment to fully integrate RT within these models.

Despite their promise, OoC platforms still face several challenges in capturing the full complexity of human organs, especially in the context of radiation biology. Progress in this field will depend on close collaboration among oncologists, radiation therapists, bioengineers, and biologists to ensure physiological and translational relevance.

Among the key challenges in applying OoC models to radiobiology are low throughput, absence of standardized protocols, small molecule absorption by PDMS, lack or limited vascularization, oversimplified microenvironmental features, and limited real-time data collection. These can be addressed by adopting modular, scalable chip designs; using radiation-compatible materials such as PMMA [82], cyclic olefin copolymer (COC) [83], or off-stoichiometry thiol–ene (OSTE) polymer [84] to avoid molecule absorption; engineering vascularized 3D cultures; incorporating ECM, immune, and stromal elements to better mimic the tumor microenvironment; and integrating biosensors (e.g., ROS, GSH) for real-time monitoring of dynamic cellular responses to RT. Moreover, incorporating on-chip fluidic control using pumpless designs or integrated micropumps could enable precise delivery of nutrients, drugs, and radiation modulators.

Developing more advanced and physiologically relevant ToC models by incorporating various TME parameters is essential for accurately predicting the effect of various radiation treatments. Most current ToC models for RT primarily utilize spheroids, organoids, or patient biopsies. Further development of these models to include tumor vascularization, immune cell circulation, fibroblasts, and gradients of oxygen and other biomolecules could significantly enhance RT research. While such capabilities have been developed for chemotherapy and immunotherapy, their application in RT could offer substantial benefits in optimizing treatment outcomes. Dual-TME OoC platforms, featuring irradiated and non-irradiated regions embedded in ECM-like hydrogels, could be used to study the migration of immune cells in response to differential chemokine and damage-associated molecular patterns (DAMPs) release. These systems simulate tumor heterogeneity and allow modeling of competitive immune cell trafficking, closely reflecting in vivo complexity. Finally, clinically-relevant biomedical imaging endpoints based on molecular contrast agents could be incorporated into ToC/OoC trials to bridge the gap between preclinical and clinical research [85]. Furthermore, ToC models could help identify optimal regimens for combination therapies, including radio-immunotherapy and radio-chemotherapy.

Developing multi-organ-on-a-chip platforms is key to understanding inter-organ communication and off-target effects, particularly by linking tumor models with radiosensitive OoCs such as heart or lungs. To increase translational value, these systems should enable long-term monitoring through embedded sensors, sampling ports, and integrated fluid handling. Clinical relevance can be further strengthened by patient-derived cells and AI-driven analytics. Expanding OoC applications to include unconventional radiotherapy modalities and radiotheranostics will further support precision radiotherapy research.

Advancing the use of OoC and ToC models—while addressing key technical challenges—will be essential for identifying countermeasures to radiation-induced injuries and enabling robust preclinical testing of radiomodulators and targeted radiopharmaceuticals. The ability to precisely control tumor microenvironmental factors on-chip is critical for accurately modeling RT responses and exploring complex treatment dynamics (see Outstanding Questions).

Outstanding questions.

Can we create translational tumor-on-a-chip models for optimizing radiation therapy for humans?

Can we use organ-on-a-chip models to identify new countermeasures to deploy after radiological disasters?

How can we control TME parameters on chip to accurately model radiotherapy responses in the in vitro setting?

Can we test and monitor clinical radiomodulating drugs and radiotracers using organ-on-a-chip models before introducing them to humans?

In conclusion, OoC platforms represent a transformative approach in radiobiology by offering human-relevant, physiologically dynamic systems for radiation research and therapeutic screening. Expanding their use in preclinical settings could significantly accelerate translation into clinically relevant radiotherapy applications and improve the development of personalized treatment strategies.

Highlights.

Moving from animal models for radiation research to organ-on-chip models would accelerate radiotherapy research and will fulfill the 3Rs principles to Reduce, Replace, and Refine animal models.

These models can be used to study both tumor responses to improve radiotherapy efficacy, and normal tissue toxicity to mitigate the harmful effects of accidental and collateral radiation exposure.

Organ-on-chip models can be derived from patient samples to individualize radiation dose prescription, considering the intrinsic sensitivity of the tumor and normal tissues.

The tumor microenvironment has a significant impact on the response of the tumor to radiation therapy. Tumor-on-chip models provide a reliable platform to control tumor microenvironment parameters and study their impact on radiation response.

Glossary

3D printing

An additive manufacturing technique used to create three-dimensional models by layering materials based on a digital design.

Abscopal effects

A phenomenon where localized radiation therapy leads to reduction of both the targeted tumor and distant, non-irradiated tumors, typically through immune system activation.

Biomaterials

Engineered substances used to interact with biological systems for therapeutic or diagnostic purposes.

Brachytherapy (BT)

an internal radiation treatment that targets cancer cells by placing radioactive material directly in or near the tumor, reducing harm to surrounding healthy tissue.

Dosimetry

The measurement of the dose (amount) of radiation absorbed by an object.

Extracellular matrix (ECM)

A network of proteins and carbohydrates surrounding cells, providing structural and biochemical support.

Glutathione (GSH)

A key antioxidant that protects cells from radiation-induced oxidative stress and damage.

Gray (Gy)

The unit of measurement for absorbed dose of radiation, equal to one joule per kilogram.

Hydrogel

A network of hydrophilic polymers that can absorb and retain large amounts of water, often used as scaffolds for cell cultures due to its tissue-like properties and biocompatibility.

Hypoxia

A condition characterized by insufficient oxygen levels in tissues or cells.

Lactate dehydrogenase (LDH)

An enzyme involved in energy production, whose presence in the extracellular space indicates cell damage.

Laminar flow

A flow regime in microfluidics characterized by smooth, parallel layers of fluid moving with minimal mixing or turbulence.

Linear accelerator (LINAC)

A device that delivers high-energy X-rays or electrons to tumors for external beam radiotherapy.

Microfabrication

Involves techniques such as photolithography, micromachining, and 3D printing to create microchannels and microscale structures for applications including organ-on-a-chip and lab-on-a-chip systems.

Microfluidics

The science of manipulating small fluid volumes via microscale channels.

Micromachining

A technique used to fabricate microscale features by removing material through processes like cutting, drilling, or etching.

Microphysiological systems

Miniaturized, engineered platforms that mimic the structure and function of human tissues or organs to model biological responses in vitro.

Multi-organ-on-a-chip

A microfluidic platform that connects multiple tissue or organ models to simulate inter-organ interactions in a physiologically relevant environment.

Photolithography

A microfabrication technique that uses UV light to expose photoresist resin, transferring a pattern onto a silicon wafer, which serves as a master mold for microfluidic chip production.

Polydimethylsiloxane (PDMS)

A flexible, biocompatible silicone-based polymer widely used in microfluidics for fabricating devices and molds due to its transparency, ease of molding, and chemical stability.

Radioresistance

The ability of cancer cells to withstand radiation therapy and continue growing.

Radiosensitizer

A substance that enhances the sensitivity of cancer cells to radiation therapy.

Radiotoxicity

The harmful effects of radiation on biological tissues.

Reactive oxygen species (ROS)

Highly reactive oxygen-containing molecules that can cause oxidative damage to cells and tissues.

Shear stress

The force per unit area applied parallel to the surface of cells within microfluidic channels.

Soft lithography

A fabrication technique that uses elastomeric stamps to create microstructures on various substrates through methods like printing, molding, and embossing.

Soft-tissue sarcoma (STS)

is a rare form of cancer that is typically treated by surgery with addition of radiotherapy

Spheroids

Three-dimensional cell aggregates that mimic the structure and function of tissues, commonly used in research to model tumors, tissue development, and drug responses.

Tissue engineering

Combining scaffolds, cells, and biochemical signals to regenerate or replace damaged tissues.

Tumor microenvironment (TME)

Complex environment surrounding a tumor, consisting of blood vessels, immune cells, fibroblasts, extracellular matrix, oxygen, nutrients, and signaling molecules, which all influence tumor progression and therapy response.