NCPI FHIR Implementation Guide - Local Development build (v0.2.0). See the Directory of published versions

Childhood Cancer

Childhood Cancer

Childhood cancers are considered rare diseases in the general population, but they are a leading cause of death in children. The underlying causes of many childhood cancers still remain unknown. Therefore, robust abilities to capture, share and aggregate data about childhood cancer is important to improve understanding. This includes overlap with rare disease research, for example there are known associations with birth defects and childhood cancer risk (https://pubmed.ncbi.nlm.nih.gov/31219523/). This helps drive the use case of reusing the NCPI Phenotypic Feature profile and the NCPI Family Relationship in the context of childhood cancer can provide the ability to aggregate and cross-analyze data for discovery.

Other data modalities that are important include pathology, imaging, and molecular characterization are critical for the research community. As pathology and imaging are already part of standard clinical care practice, the associated FHIR resources can be reused for research purposes. This could provide a streamlined way for consented data to flow from clinical systems to research platforms. The molecular characterization is a key use case across the NCPI platforms as for research purposes access to the raw data is often important for new methods development and re-analysis. This also highlights how FHIR can work with GA4GH standards as the NCPI IG provides the DRS Document Reference which is the recommended structure for pointing to a GA4GH DRS location for retrieve of the raw data, such as CRAM/BAM, FASTQ, VCF, etc.

Additionally, research on improving outcomes in childhood cancer relies on capturing treatment information, adverse events and other follow up. The longitudinal nature of the data is one of many driving use cases for the use of the relativeDateTime extension. Especially in childhood cancer, the unit of time actually captured can vary between days, months and years depending on the specific cancer type of study. While generally actual age is preferred, sometimes the deidentified time unit received is from the time of study/trial enrollment. The extension allows both the units and the value to which the time is relative to (e.g. date of birth or date of enrollment) to be flexibly but explicitly stated so that downstream systems/analysis can utilize or perform transform as needed.