Skip to main content

Conference Program

Subpage Hero

Loading

Scalable Classification of Rare Diseases Across Research Studies

Laboratory Genetics and Genomics
  • Primary Categories:
    • Genomic Medicine
  • Secondary Categories:
    • Genomic Medicine
Introduction:
The Children’s Rare Disease Collaborative (CRDC) at Boston Children’s Hospital (BCH) is a hospital-wide data repository and platform which standardizes integration of research genomic, and clinical phenotypic data for 57 rare disease groups to facilitate molecular diagnoses, precision medicine, and global efforts in advancing genomics and pediatric care. To date, 13,380 research participants have been enrolled, 3,994 exomes and 1,380 genomes have been sequenced, 15% of cases have confirmed diagnoses, and 20% of cases have candidate findings. 



The estimated number of rare diseases globally ranges from over 7,000 to 10,000, of which approximately 70% have a genetic origin. Confirming molecular diagnoses for rare diseases is limited by barriers in access, technology, and genomic knowledge, resulting in millions of individuals with rare diseases remaining undiagnosed. Several ontologies describing rare disease concepts have been developed for different purposes. While the Human Phenotype Ontology (HPO) is useful for enumerating individual phenotypic features, it is not designed to capture the full and comprehensive diagnosis. This poses the question of whether there exists an ontology that can achieve consensus for characterizing rare diseases in a unified diagnostic code. Here, we aim to describe the rare diseases within the CRDC in a systematic, meaningful way for clinical, research, and patient-care objectives.

Methods:
Investigators and their teams participating in the CRDC assigned enrolled participants disease descriptors using Mondo Disease Ontology (Mondo), a hierarchical global disease terminology system. Investigators assigned participants Mondo terms based on medical history, family history, various disease ontologies, and clinical diagnostic tools including laboratories, imaging, and genetic testing.

Results:
Over 150 rare diseases across 24 CRDC research groups have been identified using Mondo. There was variation across groups in their use of Mondo, ranging from broader disease terms, to clinical subtypes, or gene-specific diagnoses. Nineteen groups, including Cerebral Palsy, assigned singular, mutually exclusive Mondo terms for participants with and without a molecular diagnosis. Four groups utilized multiple Mondo terms to accurately describe the diseases for each participant, including Complex Fetal Anomalies and Ultra Rare Diseases. Mondo was limited or missing key disease concepts for 5 groups. Two groups reported difficulties within their specialty in describing disease subtypes. Generally, groups were able to assign gene-specific diagnoses for participants with molecular diagnoses and investigators supported and agreed with the benefits of using Mondo.

Conclusion:
Use of Mondo promotes more specific diagnosis of participants with rare diseases across the CRDC, regardless of a confirmed molecular diagnosis. Mondo’s hierarchical mapping to gene-specific diagnoses allows both bottom-up and top-down processing for understanding physician-scientists’ differential diagnoses for thousands of rare diseases. Therefore, Mondo is a powerful intermediary for rare diseases and genomics. Not only will the employment of Mondo contribute to harmonization of phenotyping within the hospital, it additionally potentiates global rare disease initiatives to advance efforts in research, diagnoses, treatment, and patient advocacy for rare diseases. These initiatives rely on unified, systematic workflows for achieving molecular diagnoses and novel gene discovery. However, challenges remain in utilizing Mondo to describe rare diseases. For some conditions it is insufficient to assign a singular Mondo label to participants, whereas for other participants, Mondo is lacking crucial disease concepts. Future steps include direct contribution of terms from disease experts to the Mondo initiative. We anticipate the addition of several hundred rare disease terms to describe all participants within the CRDC. Additionally, we expect Mondo terms for undiagnosed participants to change over time as they undergo various genomic studies in pursuit of their genetic diagnoses. Active engagement and feedback from investigators is vital for achieving consensus in rare disease ontology and improving diagnosis of unexplained rare diseases.

Agenda

Sponsors