For nearly a decade, the Cancer Dependency Map (DepMap) at the Broad Institute of MIT and Harvard has served as the "periodic table" of oncology. By systematically cataloging the genetic vulnerabilities of thousands of cancer cell lines, the project has provided the pharmaceutical industry and academic researchers with a blueprint for developing next-generation precision medicines.
However, there has long been a "flatness" problem in cancer research. For decades, the gold standard for laboratory study has been the 2D cell line—cancer cells grown on the flat, plastic surface of a petri dish. While useful, these models often fail to capture the complex, architecture-dependent behavior of real-world tumors.
In a landmark expansion, the DepMap team has integrated nearly 150 3D cancer models—organoids and spheroids—across 10 distinct cancer types. This evolution, detailed in a new study published in Nature, marks a critical turning point in how scientists identify the "Achilles’ heels" of aggressive malignancies, promising a new era of personalized cancer therapy.
The Chronology of a Scientific Milestone
The journey toward a multidimensional dependency map began in 2018 with the formal launch of the DepMap Consortium. Conceived as an academic-industry partnership, the project sought to move beyond sporadic, uncoordinated drug discovery efforts. Instead, it aimed to perform systematic, genome-wide CRISPR screens to identify which genes are essential for the survival of specific cancer cells.
The 2D Era (2018–2023)
For the first five years, the focus was primarily on traditional, immortalized cell lines. By testing over a thousand of these models, the Broad Institute created a high-resolution map of genetic dependencies. This data has already directly contributed to the development of several drug candidates currently undergoing clinical evaluation. Yet, as the database grew, so did the realization of its limitations. Certain cancer subtypes, particularly those derived from the brain and gastrointestinal tract, simply refused to grow in a 2D environment. They required the structural cues of 3D scaffolds to retain their biological relevance.
The 3D Expansion (2024–2026)
Recognizing that a "flat" map could not fully capture the complexity of human tumors, the DepMap team embarked on a massive effort to curate and characterize 3D models. They reached out to global collaborators and commercial suppliers to aggregate hundreds of patient-derived organoids—small, dome-like structures grown in specialized gels, or liquid-suspended neurospheres.
The culmination of this effort, published in the March 2026 issue of Nature, represents the most comprehensive integration of 3D and 2D cancer data to date. This work did not happen in isolation; it was accompanied by two companion studies—one led by the National Cancer Institute’s Human Cancer Models Initiative and the Dana-Farber Cancer Institute, and another by the Wellcome Sanger Institute—marking a coordinated, global push to standardize 3D cancer modeling.
Supporting Data: Why 3D Changes Everything
The integration of 3D models into the DepMap is not merely an exercise in variety; it is a fundamental shift in resolution. By comparing 2D and 3D cultures side-by-side, researchers have uncovered "blind spots" that have hindered drug development for years.
Unmasking Hidden Dependencies
The study identified specific genetic dependencies that are entirely absent in 2D models. For instance, the researchers observed that 3D glioblastoma models—a notoriously difficult brain cancer—exhibited a heightened sensitivity to CDK6 inhibition, specifically when the tumor-suppressor gene CDKN2A was missing. This dependency was obscured in 2D cultures, suggesting that the spatial orientation of the cells in 3D is required for this specific genetic pathway to manifest.
Transcriptional States and WNT Signaling
In the realm of gastrointestinal cancers, particularly pancreatic cancer, the team identified a specific gene expression program that mirrors clinical reality. This program, which is lost when cells are spread thin on plastic, remains active in 3D organoids. These organoids showed a distinct reliance on WNT signaling genes. This finding provides a direct therapeutic target that researchers previously could not have validated using traditional models.
The Variable of Culture Conditions
The research also provided a masterclass in experimental biology. The team mapped differences in cell behavior to two primary variables:
- Growth Format: Genes involved in cell adhesion and the cytoskeleton were highly sensitive to whether the cells were grown in 3D or 2D.
- Culture Medium: Genes related to lipid metabolism were driven primarily by the chemical composition of the liquid medium, regardless of the physical shape of the culture.
These findings allow researchers to choose their model with greater precision, knowing exactly which biological processes will be accurately represented by their chosen method.
Official Responses: A Vision for the Future
"These models let us see biology that we couldn’t see before," said Francisca Vazquez, Director of the Cancer Dependency Map and co-senior author of the study. "That is the whole point of DepMap. The more aspects of each tumor you capture, the more of its vulnerabilities you can catch."
Vazquez, who led the work alongside senior research scientist Tsukasa Shibue and a dedicated team including James Neiswender, Samuel Maffa, and Lisa Brenan, emphasizes that this is not a story of "replacement." Instead, it is one of "synergy."
"There is a misconception that 3D models are ‘better’ than 2D," Vazquez noted. "In our data, we see that some breast cancer markers are only present in 2D models. If we abandoned 2D, we would lose those insights. The goal is to create a complete, integrated library where the researcher understands that the map is a toolset, not a single-solution answer."
The collaboration with the National Cancer Institute (NCI) and the Wellcome Sanger Institute reflects a growing consensus in the field: the future of cancer medicine depends on the standardization of these complex models. By aligning their data structures, these institutions are ensuring that a researcher in London, Bethesda, or Cambridge is speaking the same "language" of cancer dependency.
Implications: The Road to Precision Medicine
The implications of this expanded DepMap for the pharmaceutical industry are profound.
Reducing Clinical Trial Failure
The primary reason for failure in oncology clinical trials is a lack of efficacy. By testing drugs against a broader, more representative library of 3D models, pharmaceutical companies can perform "virtual clinical trials" before a single patient is enrolled. If a drug is only effective in a specific transcriptional state found in 3D models, researchers can now identify this before moving to human trials, saving years of effort and millions of dollars.
The "Next-Generation" Patient Profile
For clinicians, this research translates to better patient stratification. The glioblastoma example regarding CDKN2A status is a perfect blueprint: a simple genomic test for a gene deletion could eventually dictate whether a patient receives a CDK6 inhibitor. This is the definition of precision oncology—using the dependency map to match the right patient with the right drug at the right time.
A Living, Growing Map
The DepMap portal is now open to the global scientific community. The team at the Broad Institute has made it clear that this is a "living" resource. As new techniques in CRISPR screening, single-cell sequencing, and organoid engineering emerge, the map will continue to evolve.
The integration of 3D models marks the end of the "flat" era of cancer biology. By embracing the complexity of the third dimension, the scientific community is moving closer to a comprehensive understanding of cancer’s vulnerabilities. While the path to curing every cancer remains long, the map is finally beginning to show the full terrain, providing the coordinates necessary for the next generation of life-saving breakthroughs.
Ultimately, the message from the Broad Institute is clear: to conquer the tumor, one must first understand the environment in which it survives. With the inclusion of these 3D models, the map has never been more accurate, more nuanced, or more ready to guide the future of cancer medicine.
About the Author
