From Living Banks to Biomarkers: Integrating PDX and PDO Data Infrastructure for Precision Oncology

Patient-derived xenograft (PDX) and organoid (PDO) models are powerful platforms for studying tumor biology and therapeutic response, but their scientific impact depends on the ability to systematically integrate and analyze the diverse data generated around them. As living biobanks grow in scale and complexity, researchers face increasing challenges related to data heterogeneity, reproducibility, discoverability, and cross-study comparability.

In this presentation, I will describe an integrated data ecosystem developed within the Princess Margaret Living Biobank to support translational research using PDX and PDO models. I will first introduce the types of clinical, molecular, and pharmacological data associated with patient-derived models and discuss key challenges encountered when scaling these resources. I will then outline a federated data infrastructure that harmonizes clinical metadata, multi-omics datasets, and drug screening results, supported by standardized analysis pipelines, shared code repositories, and user-facing query tools.

Finally, I will present a representative case study demonstrating how standardized data curation and integrated infrastructure enable efficient cohort assembly, robust analysis, and the identification of molecular determinants of drug response. This work illustrates how coupling biological insight with scalable data systems can accelerate biomarker discovery and improve the translational value of living biobank resources in precision oncology.
 

Learning Objectives:

1. Describe the major data types associated with PDX and PDO models and the challenges of managing and analyzing these data at scale.

2. Explain how integrated data curation and harmonization enable reproducible and comparable cancer research

3. Apply data-driven analyses using patient-derived models to inform therapeutic sensitivity and translational decision-making.