Patient-derived xenograft (PDX) models enhance predictability in oncology preclinical studies. Never established in vitro, PDX closely recapitulate original tumors, including histo- and molecular pathology, driver mutations, and oncogenic changes. This provides highly predictive models, correlating well with patient response to SoC, to inform on clinical study design and patient stratification.
They are useful across a range of cancer types, including those which are highly heterogeneous such as breast cancer. A panel of PDX models can represent the many different genotypic and phenotypic backgrounds seen in the patient population, and help to stratify exactly which breast cancer subtype an agent should be targeting. They are also useful in developing agents against metastatic disease.
Here’s what to look out for in a good breast cancer PDX panel, and how different features can be helpful during preclinical programs.
A Varied Breast Cancer Patient-Derived Xenograft Collection is Required
Breast cancer classification based on types and stages of the disease is complex, but essentially breaks down to three main determinants:
- Hormone driven breast cancer (ER+)
- HER2 positive breast cancer
- Triple negative breast cancer (TNBC)
A PDX collection containing hormone dependent (both ER+/PR+ and ER+/PR- status), hormone dependent and independent HER2+ (HER2+/ER+ and HER2+/ER+) and TNBC models should allow for investigation of a variety of subgroup targeted agents as well as more general breast cancer agents and their potential stratification.
Well-Characterized Breast Cancer PDX Models Can Inform on Targeted Agent Development
Characterization by RNAseq and WES allows a thorough understanding of the genetic background of breast cancer PDX models, and their use as tools in developing targeted therapies and progressing personalized medicine. BRCA1 and BRCA2 status is an important factor in breast cancer, particularly impacting on treatment with agents such as PARP inhibitors. All PDX collections should be well-profiled for gene expression, coy number, mutations, fusions etc, allowing either searching of specific models to trial new therapies based on their genetics, or for correlation of a panel response to a new agent, for developing biomarkers.
Combining this information with standard of care and experimental treatment data, can provide a well-characterized panel of breast cancer models, representing patient tumors with a high fidelity, and enabling researchers to truly understand clinical tumor growth and response in a preclinical setting. Response or resistance phenotypes to drugs such as antihormonal agents (e.g. fulvestrant), targeted agents (e.g. tamoxifen, Herceptin), and chemotherapeutics (e.g. cisplatin, taxanes, vinorelbine) can be used when choosing models to trial second or third line agents, or investigating mechanisms of genetic resistance and how to overcome them.
Model Validation is Always Needed
For every type of PDX models used, stringent validation and quality control is important, usually via pathology QC to confirm cancer type, genetic fingerprinting by STR genotyping, and also monitoring robust and consistent growth over multiple passages.
While this is not an exhaustive list of the key features of breast cancer PDX, it should help as a starting point for anyone new to the field looking to understand why breast cancer PDX are a highly useful preclinical tool.
