New information standard announced for PDX models to improve cancer research

New minimal information standard makes patient-derived xenograft (PDX) models easy to find, access, integrate and reuse

Cancer research will benefit from improved data sharing and reproducibility thanks to a new, minimal information standard for Patient Derived Xenograft (PDX) models, which are heavily used in research.

The new standard, announced recently in Cancer Research, was led by researchers at the European Bioinformatics Institute (EMBL-EBI), The Jackson Laboratory (JAX), and the University of California-San Francisco School of Medicine, in consultation with over 30 other research organizations throughout the world.

To study cancer, drug response, and advance personalized medicine, researchers explore specific biological processes using mouse models. PDX models come closer to imitating a human response to drugs than previous models by implanting human tumors within a mouse host. Because of this, laboratories around the world are increasingly using PDX models.

Without a central repository for storing PDX models, this work is being carried out in isolation, which impacts efficiency as well as the speed of research. To make such a repository possible, standards are needed so that the data going into it is consistent. Uniform information about how PDX models are generated, what quality assurance measures are applied and how the models are used helps researchers access the models made by others, and reuse them in new drug safety and efficacy studies.

“This is a very exciting step forward in our quest to study and treat cancer more effectively around the world,” said Carol Bult, scientific director of the PDX Resource at The Jackson Laboratory.  “Accelerating preclinical discoveries using PDX models hinges on the ability to meaningfully integrate data generated for PDX models by different groups. Making this integration possible was a driving force behind the development of the PDX-MI data standards.”

Researchers create PDX models by implanting tumor cells from patients with cancer into mice with suppressed immune function. The tumor that develops in the mouse reflects a patient’s own tumor properties. This allows the researcher to study a specific patient’s tumor growth and drug resistance in the mouse. There is growing evidence to show that the tumors in these mice respond — or don’t respond — to drugs in a manner similar to the patients, which is a distinct advantage over traditional cell lines and tissue-culture models.

“Unless it follows a standard, a PDX model can only help one patient or one research facility,” explains Terry Meehan, Mouse Informatics Coordinator at EMBL-EBI and principal author. “For PDX models to make a real difference in the fight against cancer, and to reduce the number of animals used in research, we need to make existing models easy to find and access. This will accelerate the rate of scientific discovery and reduce unnecessary duplication.”

“This is clearly an important standard for the community,” adds Nathalie Conte, bioinformatician at EMBL-EBI and paper author. “We sent our minimal-information standard to research institutes around the globe for their input, and the response was fantastic. In the end, we all agreed on around 40 attributes, like the age, sex, patient diagnosis, treatment response and the way the model was created. This basic information will make PDX models and their data easier to find and reuse.”

EMBL-EBI and The Jackson Laboratory are working towards the launch of an open catalogue for PDX models and associated data. These efforts, funded by the National Cancer Institute, will culminate in a shared resource hosted at EMBL-EBI.