For decades, biological research relied almost exclusively on flat plastic petri dishes. However, cells in the human body do not live in a two-dimensional monolayer. When forced onto flat surfaces, cells lose their natural polarization, experience artificial mechanical stress, and lack the complex spatial gradients of nutrients and signaling molecules found in vivo.
To bridge this translational gap, modern biomedical research has shifted toward three-dimensional (3D) patient-derived organoids, multicellular spheroids, and ex vivo organotypic cultures. These models succeed because they restore the physical, mechanical, and cellular microenvironment that dictates how tissues function, how diseases progress, and how drugs perform in the real world.
Replicating Spatial Architecture and Diffusion Gradients
In a solid tumor or healthy organ, cells are arranged in complex 3D architectures where access to oxygen, nutrients, and signaling molecules is governed by diffusion limits. A flat 2D culture exposes every cell to identical concentrations of media. In contrast, 3D spheroids and organoids naturally develop a necrotic or hypoxic core and an actively proliferating outer shell, mimicking real tissue pathology.
This spatial heterogeneity is critical for understanding disease progression and cellular identity. For instance, Natalie Landon‐Brace developed a 3D "tissue roll" model called TRACER to demonstrate that controlled radial gradients of oxygen directly drive pancreatic cancer cells to transition into aggressive, treatment-resistant basal-like subtypes (Analysis of an engineered organoid model of pancreatic cancer identifies hypoxia as a contributing factor in determining transcriptional subtypes).
Structural morphology itself also dictates drug vulnerability. Aristeidis Papargyriou et al. demonstrated that branched-organoid models of pancreatic ductal adenocarcinoma capture distinct cellular shapes and phenotypic plasticities that are entirely absent in standard spherical cultures, revealing state-targeted therapeutic vulnerabilities (Heterogeneity-driven phenotypic plasticity and treatment response in branched-organoid models of pancreatic ductal adenocarcinoma).
Furthermore, these physical barriers are essential for evaluating immunotherapies. As Marshall Chao showed, 3D tumor spheroids provide a realistic physical obstacle course to evaluate how immune cells penetrate and interact with the spatial architecture of a solid tumor, offering a more predictive platform for cell therapies (177 Preclinical characterization of CYT-100 iPSC derived NK cells alone and in combination with CYT-303 NK cell engager for hepatocellular carcinoma (HCC)).
Restoring the Extracellular Matrix and Mechanical Cues
Cells are embedded within an extracellular matrix (ECM) that provides vital mechanical and biochemical signals. Forcing cells onto rigid plastic alters integrin signaling and suppresses native behaviors. By utilizing natural or biomimetic scaffolds, researchers prompt cells to self-assemble into complex organotypic structures.
Wei‐Dong Lü demonstrated that stripping cells away from actual tumors while preserving the native acellular ECM creates a 3D scaffold where seeded cancer cells spontaneously form multicellular tumor spheroids that closely mimic native solid tumors (Development of an Acellular Tumor Extracellular Matrix as a Three-Dimensional Scaffold for Tumor Engineering). Similarly, Liang Zhang et al. utilized a tissue-derived decellularized thyroid scaffold to replicate realistic cell-to-matrix interactions, driving increased aggressiveness and drug resistance in thyroid cancer cells (Development of a biomimetic thyroid acellular scaffold as a 3D platform for modeling thyroid cancer aggressiveness and drug resistance).
When modeling complex stroma-mediated drug resistance, the exact composition of the scaffold matters. Julia Jabs engineered a composite 3D environment mixing Matrigel and Collagen I to simulate the physical density and structural complexity of the pancreatic tumor microenvironment, revealing how patient-matched fibroblasts actively shield organoids from chemotherapy by triggering a molecular identity shift (Patient-specific modeling of stroma-mediated chemoresistance of pancreatic cancer using a three-dimensional organoid-fibroblast co-culture system).
Ex Vivo Explants and Organotypic Slices
While organoids built from dissociated cells offer excellent scalability, they can lose some of the native tissue's complex architecture and stromal heterogeneity during chemical or mechanical breakdown. To preserve the highest possible fidelity, researchers leverage ex vivo explants and organotypic slice cultures, maintaining the tissue's original structural integrity without enzymatic dissociation.
Anna Golebiewska et al. demonstrated that culturing intact tumor fragments as 3D organoids preserves the original cellular heterogeneity and spatial architecture of primary and recurrent gliomas, serving as high-fidelity patient avatars for precision oncology (Patient-derived organoids and orthotopic xenografts of primary and recurrent gliomas represent relevant patient avatars for precision oncology).
This structural preservation is crucial for accurate pharmacological testing. Valeria Pecci et al. utilized ex vivo organotypic slice cultures to evaluate epigenetic drug responses in prostate cancer while keeping the native 3D microenvironment and cell-cell interactions intact (Targeting of H19/cell adhesion molecules circuitry by GSK-J4 epidrug inhibits metastatic progression in prostate cancer). Similarly, Amar Mirza et al. used primary human tissue biopsies maintained as ex vivo explants to validate synergistic therapeutic strategies for basal cell carcinoma, ensuring that the original tumor architecture and microenvironment were preserved during drug exposure (Combined inhibition of atypical PKC and histone deacetylase 1 is cooperative in basal cell carcinoma treatment).
Microfluidic Biochips and Multi-Lineage Co-Cultures
The final layer of fidelity involves modeling the dynamic, multi-lineage interactions and fluid forces present in living organs. Microfluidic "organ-on-a-chip" platforms allow for the controlled co-culture of distinct cell types under physiological flow conditions.
To study complex neurodegenerative processes, Heejung Chun et al. highlighted the use of 3D human-on-a-chip models that co-culture neurons, astrocytes, and microglia within an extracellular matrix, allowing for the direct observation of the bidirectional communication driving neuroinflammation (Elucidating the Interactive Roles of Glia in Alzheimer's Disease Using Established and Newly Developed Experimental Models).
Microfluidics also enable the high-throughput, reproducible generation of these models. Neda Azizipour et al. designed surface-optimized microfluidic biochips to generate uniform 3D tumor spheroids, demonstrating that these chips yield drug resistance patterns that closely mimic in vivo behavior compared to traditional flat cultures (Uniform Tumor Spheroids on Surface-Optimized Microfluidic Biochips for Reproducible Drug Screening and Personalized Medicine). Furthermore, these microfluidic environments can capture temporal dynamics; Jiaqi Zhou et al. developed droplet-engineered primary liver organoids that exhibit enhanced circadian rhythms, providing a robust platform for evaluating time-dependent drug toxicity (Chronotoxici‐Plate Containing Droplet‐Engineered Rhythmic Liver Organoids for Drug Toxicity Evaluation).