In simple terms, cancer organoids are organoids that are generated from cells donated by cancer patients. In our previous post about organoids, we described their many uses and applications, ranging from disease models, drug and toxicity testing, tissue and organ regeneration, and basic research to improve our understanding of biological processes such as those that govern embryonic development.

When we shift our focus to cancer organoids, it quickly becomes evident that their main applications lie in their potential to shed light on the processes of cancer development and metastasis, to help us understand heterogeneity within tumors via single cell sequencing, and to direct clinicians towards personalized cancer treatments based on patient-specific drug testing.

Lack of Clinical Translation with Conventional Cancer Models

Although our knowledge about cancer has improved significantly over the last decades, and despite significant advances in cancer screening programs, accurate diagnostics (aided by genome sequencing technologies), and the emergence of new and more targeted chemotherapeutic drugs with improved efficacy and safety profiles, cancer continues to be a major health problem worldwide, and finding the right treatment for a given patient is an everyday challenge for cancer clinicians.

Besides the challenges common to most drug discovery and development programs, the advancement of new chemotherapeutic drugs suffers particularly from poor translation from lab to clinic.

Conventional cancer models do not accurately reflect the physiology of patient tumors, and drugs that are selected as promising cancer drug candidates on the basis of rodent and xenograft models often fail in clinical trials. The clinical success of drugs that are selected in this way and perform well enough in clinical trials to gain marketing approval is also influenced by the fact that patients with the same type of cancer often respond differently to the same therapy, and resistance to chemotherapy may arise during the course of treatment.

Patient-specific tumor differences and drug resistance properties are not captured by the traditional cancer models, making it extremely difficult for clinicians to pre-emptively choose the most optimal treatment for a given patient. This may lead to loss of crucial time spent finding a suitable and safe treatment for those patients that don’t respond to the standard treatments for their cancer diagnosis.

Before the Organoid Era: Conventional Cancer Models

The most commonly used models in cancer research and drug screening are cancer cell lines and patient-derived xenograft models. Although these models have taught us a lot about cancer biology and contributed significantly to the development of many chemotherapeutic drugs, they are not without their limitations.

Cancer Cell Lines

Cancer cells can be used to generate stable cell lines to represent a plethora of different cancer types. In theory, this sounds great, but in practice, cancer cells are not always amenable to cell line establishment. Of those cancer types that do generate established cell lines, only a selection of the known subsets of any cancer type can be cultured in vitro. These two issues result in cancer cell lines that only partially represent the diversity of human cancers.

Another major drawback is that cancer cells grown in 2D culture are not surrounded by extracellular matrix (ECM), again calling into question their physiological relevance to tumors in vivo. To make matters worse, extensive passaging of cancer cell lines often results in a loss of heterogeneity as the cells gradually adapt to artificial 2D culture conditions. Passage-induced changes include altered proliferation capacities and gene expression profiles, resulting in cancer cells that move further away from biological relevance as the passage number increases.

Although methods exist to generate established cancer cell and matched healthy lines with stable karyotypes, these methods do not circumvent all of the shortcomings of cancer cell lines described above.

Patient-Derived Xenograft Models

Patient-derived xenograft models involve the implantation of pieces of tumor directly into immunocompromised mice or other animals. These models reflect in vivo tumor biology much better than in vitro cell culture, are free of the passage-induced issues associated with cell lines, and are an important component of preclinical chemotherapeutic testing, with a proven ability to predict clinical outcomes in some cases. However, they are very time-consuming to generate and validate (> 6 months on average), and not suitable for high-throughput screening or genetic manipulation. Furthermore, many of these models require larger tumor chunks than it is possible to isolate with needle biopsies.

Engraftment failure is also a serious problem for certain cancer types, such as estrogen receptor-positive breast cancer, which accounts for 70 % of all breast cancer cases. Furthermore, the possibility of animal-specific tumor evolution cannot be ruled out when using xenograft models. Collectively, these shortcomings limit the potential of patient-derived xenografts in targeted and personalized cancer treatment.

Whole-Animal Cancer Models

While whole-animal cancer models (e.g., murine) generated through manipulation of genes and pathways that are known to be implicated in cancer have provided important insights into tumor progression and metastasis, their generation is time consuming, and it is well accepted that these models don’t always accurately recapitulate pathogenic processes in patients. For example, the histological complexity and genetic heterogeneity of human cancers are typically not reflected in genetically engineered mouse models of cancer (1).

A New Era – Cancer Organoids

Cell-Line Derived Cancer Organoids

Cell-line derived cancer organoids are generated in a similar manner to healthy organoids. Typically, spheroids (multicellular 3D ball-shaped cell aggregates) derived from iPSC-derived cancer cell lines are encouraged to aggregate in specialized plastic tissue culture plates in the presence of a 3D matrix. This matrix, often matrigel, resembles the in vivo ECM, permitting self-organization into cancer organoids that mimic real-life tumors.

The 3D matrix can be further refined in order to ensure as many in vivo-like properties as possible. This might include supplementation of the 3D culture with physiologically relevant components, such as proteins, hormones, and growth factors, which are normally found in the ECM of the cancer type in question. The more that is known about the influence of ECM components on a specific tumor type, the more physiologically relevant the organoid becomes. Proteins such as collagen and laminin are only two examples of ECM proteins that might be added during this process. It is also possible to model metastasis with cancer organoids, by combining multiple tissue and tumor organoids in a single 3D platform.

Although this approach is highly useful, its efficiency depends on cancer type and the genetic landscape of that tumor, for example, whether or not it contains specific oncogenic mutations. An important risk with cell-line derived cancer organoids is that one might select for certain tumor subclones, potentially resulting in organoids that don’t reflect the genetic heterogeneity of the original tumor.

Patient-Derived Cancer Organoids – Tumor-On-A-Chip

Here, 3D tumor organoids are generated directly from freshly isolated tumor biopsies to yield patient-specific tumor models. These serve as a ‘tumor-on-a-chip’ to model a patient’s cancer outside of their body, and to pre-emptively direct oncologists towards the optimal cancer treatment for that patient.

To generate patient-derived cancer organoids, fresh tumor biopsies are used to generate viable tumor constructs with the help of advanced 3D culture technology, similar to that used for the development of healthy organoids. Organoids generated from patients in this way, supported by in vivo-like 3D culture conditions that mimic the known tumor microenvironment, provide a novel opportunity to test a range of candidate chemotherapeutic drugs in parallel and with a short turnaround time, thus saving precious time and sparing patients of unnecessary side effects associated with treatments that their cancer may never respond to. To date, long-term patient-derived cancer organoids have been established from a wide range of tissues, including colon, breast, and prostate, and studies have revealed these organoids to closely resemble the original tumors both phenotypically and genetically.

Personalized Medicine

Patient-derived cancer organoids are thought to reflect an individual’s cancer more faithfully than any other cancer model to date because they are derived directly from patients and supported by an in vivo-like 3D microenvironment. Indeed, studies using tumor-derived organoids for tumor-on-a-chip chemotherapy screening are beginning to emerge, and the results are promising (2). The technology also creates the opportunity to rapidly genetically profile tumors for patient-specific mutations that might influence the outcome of candidate treatments. Such mutations can then be exploited to find personalized cancer treatments in cases where marketed drugs that are not used in the treatment of a given cancer (or not used in cancer treatment at all) may have activity against an individual’s cancer.

The possibility to create matched pairs of tumor-derived and healthy organoids from the same individual allows screening for drugs that selectively target the cancer cells in that individual. It also creates a new avenue to further our understanding of the genetic and epigenetic contributions to cancer development, as well as resistance to chemotherapy.

The Future for Cancer Organoids

The cancer organoid field is rapidly growing, and a number of clinical research institutes and companies around the world have already started to incorporate tumor-on-a-chip technology into their cancer research and drug development pipelines. Although not discussed here, cancer organoids have also been explored in the field of immunotherapy, where under certain conditions they may boost the numbers of cytotoxic immune cells in culture to improve the efficiency of patient-derived immune cell therapies.

Ultimately, better cancer treatments will be developed using improved in vitro model systems, including patient-derived organoids as well as cancer organoids derived from iPSCs. In theory, the latter represent an unlimited source for organoid development, and this is noteworthy given that not all tumors are amenable to biopsy or in vitro culture. As time goes on, we may see the applications for cancer organoids in every facet of cancer research and treatment, from increasing our understanding of cancer biology right through to directing clinical decisions made by oncologists in cancer clinics.


  1. Cheon DJ, Orsulic S. Mouse models of cancer. Annu Rev Pathol. 2011;6:95-119.
  2. Mazzocchi AR, Rajan SAP, Votanopoulos KI, Hall AR, Skardal A. In vitro patient-derived 3D mesothelioma tumor organoids facilitate patient-centric therapeutic screening. Sci Rep. 2018;8(1):2886.

Further Reading:

Drost J, Clevers H. Organoids in cancer research. Nat Rev Cancer. 2018.

Article by Karen O’Hanlon Cohrt PhD. Contact Karen at 

Karen O’Hanlon Cohrt is a Science Writer with a PhD in biotechnology from Maynooth University, Ireland (2011). After her PhD,  Karen moved to Denmark and held postdoctoral positions in mycology and later in human cell cycle regulation, before moving to the world of drug discovery. Her broad research background provides the technical know-how to support scientists in diverse areas, and this in combination with her passion for writing helps her to keep abreast of exciting research developments as they unfold. Follow Karen on Twitter @KarenOHCohrt. Karen has been a science writer since 2014; you can find her other work on her portfolio.