A brief overview of human cancer models

Why study cancer?

There is hardly anyone on this planet who has not heard about cancer and has not been in touch with it in this one way or another.

It looks like cancer has existed for as long as humanity remembers itself; the views on causes and treatments varied and evolved but... we are not there yet! There is still a lot to discover before we understand and learn to handle cancer. I highly recommend reading this succinct and informative history of cancer.

Using models in drug discovery

Every time researchers find a (new) chemical that they suspect might have anticancer properties, they have to study it extensively before it makes it into clinical trials

Most chemicals won't even make it that far, and those that will, might take more than 10 years to reach the market and the process may cost several billion pounds. It is important to aim to reduce the costs at all stages, so they usually start from relatively cheap models and if drug passes all tests, more complex and expensive models are employed. Only if a drug has successfully passed all cell- and animal model-based experiments can it go further into clinical trials (where it can still fail!).

Cell-based assays

The first method to study effects of a drug is based on cell cultures. In this method, different cell types are used to estimate whether there is a biological response to the drug under investigation.
This is done by measuring cell count, DNA in growing cells, studying cell respiration patterns etc.
The method is easy, fast, relatively cheap, provides homogeneous population of cells. The cells are easy to manipulate, many things are known about established cell lines; the method also allows to work out molecular pathways. Disadvantages of the cell-based assays include using reductionist approach; cells growing in 2D space does not have the complexity that cells growing in vivo has (although this is partially addressed by using 3D cell cultures). Cell cultures also lack whole animal physiology, immune system, interacting organ systems, microenvironment and heterogeneity of cancer in patients. Cells in culture cannot be used to evaluate drug delivery, metabolism and toxicity.

Animal models

So, once the initial screening has demonstrated that a certain substance has an anticancer effect on cells in culture, one can move the studies in vivo.
Many animal models are used in research and animal use is heavily supervised and regulated. One always has to keep in mind that use of animals in research setting is a privilege, not a right.
The following animal models are discussed below1,2:

  • Xenograft
  • Spontaneous
  • Carcinogen-induced
  • Genetically engineered mice

Xenograft models

In PDX model3 (PDX stands for patient-derived xenograft), cancerous tissue from a patient's tumour is implanted directly into immunocompromised mice. PDX models retain tumour heterogeneity, gene expression, and response to treatment of the original tumour. PDX models are thus thought to be more translationally relevant, especially as a drug development tool, because PDXs can capture the genetic character and heterogeneity that exists within a single patient’s tumor and across a population of patients’ tumors. PDX models also hold enormous potential for identifying predictive markers for therapeutic response. But because the host's immune system is different from human and compromised, this model cannot be used to study drugs that affect immune system. It also does not recapitulate the complexity of tumour progression and one cannot test prevention or early events.

Spontaneous & carcinogen-induced models

From many years of mouse studies, it is known that some strains of laboratory animals are susceptible to spontaneously developing certain types of tumours. These models may mimic some types of human diseases, can be used to study early disease and prevention. Disadvantages include variability of disease progression, large animal numbers needed, long time to develop disease, different penetrance (not all animals will develop disease).

In carcinogen-induced cancer models an animal is treated with a carcinogen to induce cancer. Examples include lung, skin, bladder, stomach, prostate cancer. This model mimics initiation steps of some cancer, can be used to study early events, to identify predisposing conditions and to study prevention. Disadvantages include potential health hazard to investigator, variability of disease progression, and large animal numbers needed.

eMICE portal gives a brief overview of these models and the Jackson Laboratory website lists available mouse models of these types.

Genetically engineered mouse models (GEMM)

Genetically engineered mice have induced mutations, including transgenes, targeted mutations (knockouts or knockins), and retroviral, proviral or chemically induced mutations.
GEMMsdevelop de novo tumours in a natural immune‐proficient microenvironment. Tumours arising in advanced GEMMs may closely mimic the histopathological and molecular features of their human counterparts, display genetic heterogeneity, and, in some cases, are able to spontaneously progress toward metastatic disease 4.

This model may mimic initiation steps of some cancers; can be used to study early events, to test genetic lesions that predispose to cancer; it is autochthonous (rising in the tissue of origin); penetrance is usually 100%; immune system stays intact; disease can progress with time. Disadvantages include variability of disease progression; requirement of large animal numbers; disease initiator may be artificial; it is time consuming to characterise and validate the results; if integrated later in development, chimeric offsprings can be produced.

Repositive's input: making PDX models more accessible for researchers

In collaboration with AstraZeneca, Repositive has launched a PDX consortium to develop a collaborative, pre-competitive PDX resource. The objective of the project is to provide a resource for streamlined discovery and queries on molecular data from PDX models across multiple sources to suit specific needs in oncology research.

You can find out more about the PDX consortium by reading the following posts:

Finding the right PDX model

Mouse Trap: How to win in the PDX game

References (Open Access)

Read more posts by Nadia Kovalevskaya