From 2D to 3D: A Century-Long Evolution of In Vitro Tumor Models

In the protracted battle against cancer, one of the frontiers of human ingenuity is the construction of in vitro models in the laboratory that mimic real tumors. This is more than just simple cell culture; it represents a cognitive revolution, moving from two-dimensional (2D) to three-dimensional (3D), and even beyond. This revolution has profoundly altered our understanding of tumor biology and opened up unprecedented avenues for developing new therapeutic strategies.

The Birth and Limitations of a "Map"

The story begins in the early 20th century, when scientists Alexis Carrel and Montrose Burrows first successfully kept tissues and cells alive in vitro. This was a groundbreaking achievement at the time, but these primary cultured cells were like fish out of water: short-lived and quickly senescent and apoptotic, making long-term, systematic research virtually impossible.

The turning point came in 1951, when the HeLa cell line was successfully established. HeLa cells, capable of proliferation indefinitely in the lab, provided a stable, reproducible tool for cancer research. Their emergence was like the first reusable "map" for cancer research, allowing investigators to delve deeper into cancer at the molecular, cellular, and genetic levels than ever before.

The success of HeLa cells ushered in a "golden age" of cell line development. By the 1970s, more iconic tumor cell lines, such as the breast cancer cell lines MCF-7 and MDA-MB-231, were developed. These two-dimensional cells, growing flat on the bottom of a plastic culture dish, formed a flat "cell society". Their uniformity made them easy to observe and manipulate, becoming the "standard model" for cancer research.

However, this "map" was ultimately flat. A real tumor is far more complex than a single, flat layer of cells can capture. Tumors within the human body are a bustling, chaotic, and vibrant three-dimensional "city". It not only contains cancer cells, but also stromal cells that support their growth, immune cells involved in the immune response, and the extracellular matrix (ECM) that binds them together. More importantly, there are complex and dynamic interactions between cells. The complexity of this three-dimensional structure and microenvironment is precisely the most fatal flaw of the 2D model. Therefore, anti-cancer drugs screened based on 2D models often suffer disastrous failures after entering clinical trials because they only attack an oversimplified "virtual enemy". Researchers realized that to truly understand and defeat cancer, they must build a 3D model that can truly simulate the complexity of tumors.

From Flat to Spheres: The First Step Toward 3D Models

Recognizing the limitations of 2D models, researchers began exploring 3D models. The earliest attempt emerged in the 1970s, with the development of a model called "spheroids". Their construction is ingenious: cancer cells are seeded in non-adhesive culture dishes, such as ultra-low attachment plates or hanging drops. Unable to adhere to the bottom of the dish, the cells naturally aggregate with each other, forming a 3D cellular aggregate—the spheroid.

Compared to flat monolayers of cells, spheroids are undoubtedly a significant advancement. For the first time, they simulate certain key architectural features of solid tumors in vitro. Within a sufficiently large spheroid, cells in the outer layers enjoy ample access to nutrients and oxygen, while cells in the core suffer from insufficient supply, leading to hypoxia and nutrient depletion, and even necrosis. This outside-to-inside gradient faithfully replicates the microenvironmental gradients commonly found in solid tumors, providing a more realistic platform for studying tumor growth, pre-angiogenic states, and response to radiotherapy.

A specialized subtype of spheroids has emerged: tumorspheres. These are typically formed from cancer cells with stem cell properties in a serum-free, specialized culture medium. These tumorspheres are enriched with cancer stem cells, which are considered the root of all evil in tumor recurrence, metastasis, and drug resistance. Therefore, the tumorsphere model has played a crucial role in studying cancer stem cell biology, tumor heterogeneity, mechanisms of drug resistance, and metastatic potential.

However, both spheroids and tumorspheres lack a crucial element: the extracellular matrix (ECM). For a long time, the ECM was considered merely a "scaffolding" providing structural support. It wasn't until 1982 that researcher Mina Bissell and her colleagues proposed a groundbreaking concept: the ECM, far from being a scaffold, is a biologically active signaling center that profoundly influences gene expression and cell behavior by regulating tissue structure. This insight was a paradigm shift in cancer research. It implied that a cell's fate is determined not only by its inherent genes but also by its surroundings.

The experimental validation of this theory was aided by a key technological breakthrough in the 1980s: the development of Matrigel. Matrigel is a gel-like protein mixture extracted from mouse sarcomas and rich in ECM components such as laminin and type IV collagen. It is liquid at low temperatures and solidifies into a gel at 37°C, mimicking the structure and function of the basement membrane in vivo.

The advent of Matrigel revolutionized 3D cell culture. It provides cells with a 3D environment that is closer to their physiological state, allowing researchers to observe the behavior of cells in a more realistic setting.

The Birth of Mini-Organs and the Rise of Personalized Tumor Models

In the 21st century, with the rapid development of stem cell biology, a more advanced and biomimetic 3D model emerged: organoids. Organoids are 3D cell cultures formed by the self-organization and differentiation of stem or progenitor cells under specific in vitro conditions, reproducing the specific structure and function of their source organs. They are like "mini-organs" grown in a petri dish.

Originally, organoids were primarily derived from healthy adult stem cells, such as those isolated from intestinal, liver, or breast tissue. These normal organoids provide powerful tools for understanding tissue homeostasis, regeneration, and disease modeling.

A revolutionary breakthrough has been the development of patient-derived organoids (PDOs). As the name suggests, PDOs are derived directly from a patient's tumor tissue, which can be obtained through surgical resection, biopsy, or even fine-needle aspiration. This means that for the first time, we can create a "living" tumor model in the laboratory for each patient—a truly personalized "mini-tumor."

First, fresh tumor tissue obtained from the patient undergoes delicate mechanical disruption and enzymatic digestion to break it down into single cells or small cell clusters. Next, the cell suspension is mixed with Matrigel and carefully placed in a dome-like pattern in the center of a culture dish or evenly spread across the bottom of a well. The Matrigel solidifies at 37°C, providing the necessary structural support and biochemical signals for the cells to self-organize. Finally, a culture medium tailored to the specific tumor type is added, containing key growth factors and small molecules, such as Wnt pathway activators and epidermal growth factor (EGF), to promote cell proliferation and differentiation.

Over the next 7 to 14 days, a miracle occurs. The patient-derived cancer cells reorganize within the Matrigel, proliferate, and form tumor-like organoids with complex 3D structures. PDOs offer unparalleled advantages: they fully retain the genetic mutations, gene expression profiles, and cellular heterogeneity of the original tumor.

The success rate of establishing PDOs varies depending on tumor type. For example, the success rate for breast cancer PDOs can exceed 80%, while that for pancreatic and colorectal cancers is over 70%. However, for tumors with higher heterogeneity, slower proliferation, or those requiring a more specialized microenvironment, such as prostate cancer, the success rate can be less than 20%.

Going a step further, researchers have developed patient-derived xenograft organoids (PDxOs). This technique involves implanting patient tumor tissue into immunodeficient mice, allowing it to grow in a more physiological in vivo environment. The expanded tumors are then removed from these mice and used to establish organoids. While this process is more complex, it provides another powerful model for studying tumor progression and drug response in vivo.

The most exciting application of PDOs lies in personalized precision medicine. Researchers can test the efficacy of various chemotherapeutic agents or targeted drugs on these "mini-tumors," thereby predicting a patient's response to specific treatment options and providing valuable insights for clinicians to develop optimal treatment strategies. Furthermore, PDOs can be cryopreserved and revived, allowing for the found of a "living" biobank of patient samples for future research and therapeutic exploration.

Beyond the "Tumor City" to a Dynamic "Tumor Ecosystem"

Current organoid models typically lack key immune and stromal cells, as well as vascular networks that mimic blood flow. These elements together constitute the complex tumor microenvironment (TME), which plays a crucial role in tumor growth, invasion, and response to therapy. Furthermore, the reliance on Matrigel leads to batch-to-batch instability, limiting experimental reproducibility.

To overcome these challenges, researchers are exploring cutting-edge technologies, striving to transform static "tumor cities" into dynamic "tumor ecosystems".

1. Miniaturization and Automation: Microorganoid Spheroids (MOSs) and High-Throughput Screening

To address the time-consuming and sample-intensive culture requirements of traditional PDOs, a technology called "micro-organospheres" (MOSs) has emerged. Utilizing droplet microfluidics, MOSs can rapidly generate thousands of uniform microscopic tumor spheres from extremely small biopsy samples. This technology not only significantly improves culture efficiency and scalability, making it more suitable for large-scale drug screening, but also reportedly retains a certain degree of stromal and immune cell components from the original tumor, opening up the possibility of testing immunotherapies.

2. Introducing "Blood Flow": Tumor-on-a-Chip

To mimic the dynamic interaction between tumors and the circulatory system, researchers have combined organoid technology with microfluidics to create a "tumor-on-a-chip". This model integrates a network of microchannels within a small chip, allowing perfusion of culture medium or "artificial blood" to simulate in vivo blood flow. Tumor organoids, spheroids, or tumor tissue slices are placed in specific chambers within the chip, allowing them to experience physical signals such as fluid shear and interstitial flow that are completely absent in static culture.

This "metastasis-on-a-chip" model provides an unprecedented platform for testing anti-metastatic drugs. Furthermore, by integrating different "organ modules" (such as a liver module) on the chip, it is possible to simulate the metabolism and toxicity of drugs in vivo, creating a "multi-organ chip" system.

3. "Recruiting Residents": Co-culture Systems and Microenvironment Remodeling

To ensure that the "tumor city" is no longer isolated, researchers are actively recruiting new residents. By co-culturing tumor organoids with stromal cells (such as cancer-associated fibroblasts (CAFs)) and immune cells (such as T cells) isolated from patients, the complex cellular ecology of the tumor microenvironment can be better recreated.

A particularly ingenious technique is the "air-liquid interface" (ALI) culture. This technique culture cells in a specialized chamber (transwell) with a porous membrane. The lower half of the membrane is immersed in culture medium for nutrients, while the upper half is exposed to air. This approach mimics the physiological environment of epithelial tissue in organs such as the lungs and intestines. Recently, researchers have combined ALI technology with PDOs to successfully preserve the tumor's native T cell receptor repertoire in an in vitro model, providing a powerful tool for studying immunotherapies such as immune checkpoint inhibitors.

These cutting-edge technologies are driving the evolution of in vitro models from static, simplified cell aggregates to dynamic, multi-component, and functionally integrated "micro-ecosystems". We are one step closer to truly recreating a living, fully functional tumor in the laboratory.