Skip to Content
Patient-Derived Models Repository (PDMR)
Last Updated: 06/28/18

NCI PDMR Models Overview

Diagram of the PDMR Model Generation Workflow

Overview

The PDMRs primary source of patient material are two NCI-sponsored protocols: NCI Tissue Procurement Protocol (clincialtrials.gov: NCT00900198) and CIRB Tissue Procurement Protocol 9846. In addition to fresh tissue specimens, we have also obtained viably cryopreserved patient tumor tissue, rapid autopsy material, and previously-derived PDXs from outside groups. All such specimens have been collected under IRB-approved protocols and released through MTAs to allow use and distribution with the PDMR.

The goal is to have at least 1000 PDX models available in the PDMR. In the future, early-passage mixed tumor cultures as well as cancer-associated fibroblast cultures will be released through the PDMR for many of the same originating patients. We continue to seek external site (profit and non-profit) contributors to provide previously-derived models or patient material to the PDMR for model generation and release to the research community. All models received by the PDMR go through the same QC procedures that are applied to internally-derived models.

Quality Control

The PDMR places stringent QC criteria on models.

  • Short Tandem Repeat (STR) profiles are provided at the patient-level in the PDMR database for model validation. The STR profile for any model derivative (PDX, PDC, CAF, PDOrg) is provided for reference.
  • Human DNA content is analyzed by qRT-PCR to verify model origin.
  • Pathology and/or IHC markers are confirmed relative to the patient diagnosis with all tumor models by a clinical pathologist.
  • Concordance of STR profile and WES across all characterized model data

PDX-Specific Quality Control

Patient-Derived Xenograft (PDX): A tumor model generated by implantation of fresh human tissue (tumor, enriched CTCs) into immunodeficient mice. The PDMR host mouse model used is NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ (NSG).

  • Originator: Original patient material used for model generation
  • Passage 0 (P0): First mouse passage implanted with the original human specimen

The PDMR expands its models primarily by subcutaneous fragment engraftment to maintain as much tumor heterogeneity as possible. Distribution lots are early passage — generally P1-P4. Because of this, variations in histology and in low allele frequency genetic variants can be observed in lineages within a model. These same types of variations can be expected in PDX models expanded by the recipients of distributed fragments from the repository. While PDX models generally demonstrate a high degree of stability once they reach P3-5, the PDMR has observed some models that continue to evolve; this caveat should be considered by any investigator requesting models for their research aims.

PDX models are proven human origin, have been confirmed by histopathology to match the patient diagnosis or pathology of provided patient material, completed IDEXX human pathogen assessment, been successfully regrown from cryopreserved material, have STR profile and human DNA content validated for the entire distribution lot, and have whole exome sequence and RNASeq data available from 4-6 representative PDX models.

PDC- and CAF-Specific Quality Control

The PDMR generates in vitro models from the following material in order of priority: patient, PDX, patient-derived organoid (PDOrg); see above figure. CAFs are primarily generated from patient material, though occasionally they can be recovered from a P0/P1 PDXs. The origin of material used to generate the model is listed in the PDMR database with the model growth information. PDC and CAF cultures use specialized Complete Media and grow with a variety of growth characteristics. The model-specific Certificate of Analysis (provided with distributed material) and PDMR SOPs (https://pdmr.cancer.gov/sops/default.htm) should be used to ensure culture success — these models should not be treated like traditional in vitro cultures (e.g., HeLa, MCF7)

  • PDC - Patient-Derived Mixed Tumor Cell Cultures: 99.9% human tumor culture, non-clonal (multiple cell types or non-cloned lineages, aka “mixed”), 99.9% fibroblast free (FACs analysis). The distribution lot has also been subcutaneously implanted into NSG mice to test for tumorigenicity. These cultures may or may not be able to form spheroids in a fully-defined, serum- and feeder-free medium or in soft agar.
    • These cultures have been grown for at least 20 passages from the distribution lot and no out-growth of fibroblasts has been observed.
  • CAF — Cancer Associated Fibroblast Cultures: 99.9% human fibroblast culture and 99.9% human tumor cell free (FACs analysis). These cultures have been proven to be non-tumorigenic following subcutaneous implantation into NSG mice and have been characterized as fibroblasts by at least one of the following methods: immunohistochemistry, qRT-PCR, or FACs.
    • It is important to note these are non-transformed cells. CAFs have a finite lifespan in vitro
    • CAFs are guaranteed for experimental use for up to 3 passages when maintained on Matrigel-coated surface in the recommended Complete Media + Y compound
    • Additional population doublings and subcultures are possible, but overall fitness of culture may deteriorate with subsequent passages