NCI Cancer Model Development
NCI-supported cancer model development is a collaborative effort, which includes model development centers and model processing entities that provide clinical data, quality control (QC), biospecimen processing, and molecular characterization.
Map of the Cancer Model Development Centers (CMDCs) and model processing entities. The map shows locations of the CMDCs (purple): Broad Institute in Cambridge, Massachusetts, USA; Cold Spring Harbor Laboratory in Cold Spring Harbor, New York, USA, and their subsites: Hubrecht Institute in Utrecht, Netherlands and University of Verona in Verona, Italy; Stanford University, Stanford, CA; Weill Cornell Medical College, New York, NY; Biospecimen Processing Center (green) at the Nationwide Children’s Hospital in Columbus, Ohio, USA; Genome Characterization Centers (orange) at Broad Institute in Cambridge, Massachusetts, USA, and University of North Carolina in Chapel Hill, North Carolina, USA; Genomic Data Commons (pink) in Chicago, Illinois, USA; and Clinical Data Coordinating Center (teal) at the Information Management Services in Rockville, Maryland, USA; and web developer for HCMI Searchable Catalog at the Ontario Institute for Cancer Research (gray) in Toronto, Ontario, Canada.
The Cancer Model Development Centers (CMDCs) are the NCI-supported contributors to the HCMI. CMDCs are tasked with producing next-generation cancer models from clinical samples. The cancer models include tumor types that are rare, originate from patients from underrepresented populations, lack precision therapy, or lack cancer model tools. Throughout the development process, the CMDCs utilize stringent internal QC measures to ensure both clinical and molecular integrity. These models' case-associated clinical, biospecimen and molecular characterization data are available as a community resource.
The CMDCs are:
Broad Institute of MIT and Harvard (BROD): The Co-Directors are Jesse S. Boehm, Ph.D. and Keith L. Ligon, M.D., Ph.D.
Cold Spring Harbor Laboratory (CSHL): The Principal Investigator is David A. Tuveson, M.D., Ph.D. The CSHL CMDC includes two international subsites: 1) ARC-NET Centre for Applied Research on Cancer, University of Verona, co-led by Aldo Scarpa, M.D., Ph.D. and Vincenzo Corbo, Ph.D. and 2) Hubrecht Institute, led by Hans Clevers, M.D., Ph.D.
Stanford University: The Principal Investigator is Calvin J. Kuo, M.D., Ph.D.
Weill Cornell Medical College: The Principal Investigator is Olivier Elemento, Ph.D.
The HCMI CMDC-specific model development and characterization pipeline involves multiple institutions. It includes several QC check points which ensure that the models and case-associated clinical, biospecimen and molecular characterization data are complete and consistent.
CMDC Simplified Flowchart. The models are first generated by the CMDCs and clinical data associated with the model's case is submitted to the CDC. The CDC QC’ed clinical data for the model's case is sent to the GDC. Samples of normal tissue, parent tumor, and derived model are sent to the BPC for isolation of nucleic acids. The QC’ed nucleic acid samples are sent to the GCCs for WXS, WGS and RNA-Seq and raw sequencing data is submitted to the GDC. CMDC-validated models and their case-associated genomic and clinical data will be provided as an HCMI resource to the research community. Each component within the flowchart is hyperlinked to corresponding text below or to resources.
The clinical data is submitted to the CDC at Information Management Services Inc. The CDC collects and QCs the clinical data to confirm patient privacy is protected and to certify that the submitted clinical data conforms to the controlled vocabulary of the case report forms (CRFs). As part of the clinical data submission process to GDC, the CDC maps the clinical data to the GDC data dictionary so that users can search and filter by specific clinical data.
Tissue samples from patients and derived cancer models are sent to the BPC at Nationwide Children's Hospital. DNA is isolated from normal tissue, parent tumor, and the derived model while RNA is isolated from the parent tumor and derived model using standardized protocols. Internal BPC QC and genotyping ensure quality and consistency of material for molecular characterization. The BPC submits biospecimen data to GDC and sends isolated nucleic acids to the Genomic Characterization Centers (GCCs) for molecular characterization.
The isolated nucleic acids are sent to the GCCs for molecular characterization.
- Broad Institute: 15x Whole genome sequencing (WGS) and 150x whole exome sequencing (WXS) are performed on DNA from normal tissue, parent tumor (when available), and derived model. Infinium MethylationEPIC DNA Array is performed on DNA from the derived model and a subset of tumors.
- The University of North Carolina: 120 million read RNA sequencing (RNA-Seq) is performed on RNA from the parent tumor and the derived model.
The NCI's GDC houses all the clinical, biospecimen, and molecular characterization data. The GDC performs QC and harmonizes the sequencing data through their analytical pipeline. Harmonized data are available at the GDC data portal.
All data is made available using NIH policies that protect patient privacy and confidentiality.
HCMI Searchable Catalog is a continuously updated resource for querying the available next-generation models developed by HCMI. Within the Catalog, users can search by patient demographics, tumor, and model elements including age at diagnosis, sex, treatment information, clinical tumor diagnosis, primary site, clinical stage, and type of model (e.g. 3D-organoid, 2D-conditionally reprogrammed cells), open-access masked somatic MAF variants, etc. For additional assistance in navigating the Searchable Catalog, please see the “HCMI Searchable Catalog User Guide”.