• Home /
  • Blog /
  • Patient Derived Xenografts:  The Crucial Piece for Curing Cancer

Patient Derived Xenografts:  The Crucial Piece for Curing Cancer

Jan 4, 2017 3:57:00 PM No Comments
PDX_Patient Derived Xenografts.jpg

How to Identify an Accurate Model for Your Efficacy Study

The initial results are exciting, and now you are eager to take your compound to the next step. However, finding the best preclinical model can be challenging and even frustrating. It's even more daunting if you consider that hundreds of millions of dollars have been spent on development, and there is only a 13% chance that your compound will make it through all phases of the clinical trial into market. Here, we will talk about how to identify the best model for your in vivo efficacy study resulting in higher clinical predictability therefore reducing cost.

Assessing the Landscape: PDX Models Versus CDX Models

So you have a promising compound you would like to test against a particular cancer. The question you need to ask yourself is: "How do I choose the best model for efficacy testing"? Of course you already know the type of cancer you want to screen against, but then what? Do you choose a CDX model or a PDX model? Let's first distinguish the difference between the two. CDX models or cell line derived xenograft models, are created with an established cell line that has been adapted to grow in cultured conditions. Because the cell lines that are used for CDX models have been passaged in artificial conditions, the risk of genetic and phenotypic drift is high (Rosfjord et al. 2014). As you can imagine, this creates a model that is a very poor clinical representation. PDX models (patient derived xenografts) on the other hand, are passaged in mice directly. The tumor is excised from the patient and immediately implanted into a carrier mouse. This greatly reduces the risk of genetic drift because the artificial culturing step has been eliminated. The tumor is then expanded only in vivo.

PDX figure-1.jpg

Now that you know that PDX models are the best outlet for your efficacy test. We can now answer the question:

How do I choose the best PDX model for efficacy testing?

Below are a few helpful guidelines to use when looking for the optimal PDX model for you efficacy study:

  1. A True Representation of the Disease

    After Identifying a model that has the targeted mutation of interest you'll want to know if the model has maintained it's fidelity (in other words does it genetically and phentoypically still resemble the original patient tumor). Even though the propensity for genetic drift in PDX models is much lower than CDX models, drift does occur and it's important to validate that the model hasn't drifted. In order to do this, an indepth genomic comparison (whole genome sequencing is best) needs to be conducted to compare the parent tumor with the xenograft model. So the take home message is to choose high fidelity PDX models that have been well characterized to adequately evaluate genetic drift and to verify that the targeted mutation is still present.

  2. Have Extensive Patient History

    Having access to the patient history such as age, sex, and treatment regimen of the model you plan on using is vital. This will answer important questions such as: Is it a naive tumor that hasn't been exposed to any treatment or has the tumor been exposed to a gauntlet of various treatments? Since some tumors are affected by age and sex dependent fluctuations in hormone levels, the patient's age and sex is equally important. Without this crucial patient history information the efficacy study would be extremely difficult to interpret. Additionally, it's important to know if the treatment response in the PDX model recapitulates the treatment response of the patient tumor. Stromal differences between human and mouse species may play a role in differential drug response (Zhao et al.2012). The company you purchase the PDX model from should validate that the treatment response has or has not changed.

  3. Investigate the Company


    ThisTumorimagerTM image was used with the generous permission of Biopticon.

    The reputation of the company you choose is very important as well. What happens behind the scenes could have a major effect on the efficacy study results. Here are some questions you need to ask: Does the company follow IACUC protocols for the humane treatment of animals? Are the studies conducted in sterile conditions? The best way to answer these questions is to request an onsite visit. If the company hesitates or states that they don't allow oniste visits, that should be a major red flag. Other important questions to ask are: How is the data gathered? How are the results analyzed? What data do I receive from the study? The company should have a standard way of measuring tumor size, randomizing the experiment, and analyzing the results.


With your new knowledge and your new compound in hand, you are now ready to take it to the next level. Here is a review of how to choose the best PDX model for your efficacy study: Growth curve example-2.jpg

  • Choose a high fidelity PDX model
  • Choose a well characterized model (example: WGS and extensive patient history information)
  • Choose a company that is reputable and adheres to high standards
  • Know how the data is generated and what data you will get back


Rosfjord E, Lucas J,Li G., Gerber HP. Advances in patient-derived tumor xenografts: From target identification to predicting clinical response rates in oncology. Biochemical Pharmacology. 2014; 91:135-143.

Zhao X, Liu Z, Yu L, Zhang Y, Baxter P, Voicu H, et al. Global gene expression profiling confirms the molecular fidelity of primary tumor-based orthotopic xenograft mouse models of medulloblastoma. Neuro-oncology. 2012;14:574-83.

PDX Models Curing Cancer in 3 Easy Steps

What are your experiences with efficacy studies?

#PDX, #preclinical efficacy testing, #Cancer models, #High Fidelity Models

Spying on proteins - How observation can affect the perception of reality

One of the highest ideals in science is to observe natural events in their native context. Doing so...