A scalable EHR-based approach for phenotype discovery and variant interpretation for hereditary cancer genes

Chenjie Zeng; Lisa A. Bastarache; Ran Tao; Eric Venner; Scott Joseph Hebbring; Justin D. Andujar; Harris T. Bland; David R. Crosslin; Siddharth Pratap; Ayorinde Cooley; Jennifer A. Pacheco; Kurt Derek Christensen; Emma F. Perez; Carrie L. Blout Zawatsky; Leora Witkowski; Hana Zouk; Chunhua Weng; Kathleen A. Leppig; Patrick M. Abou‐Sleiman; Hakon H. Hakonarson; Marc S. Williams; Yuan Luo; Gail Pairitz Jarvik; Robert C. Green; Wendy K. Chung; Ali G. Gharavi; Niall J. Lennon; Heidi L. Rehm; Richard A. Gibbs; Josh F. Peterson

doi:10.1001/jamaoncol.2022.0373

Genetic & Genomic MedicineJAMA OncologyApr 21, 2022

Peer-reviewed journal article

A scalable EHR-based approach for phenotype discovery and variant interpretation for hereditary cancer genes

Chenjie Zeng iD, Lisa A. Bastarache iD, Ran Tao iD, Eric Venner iD, Scott Joseph Hebbring iD, Justin D. Andujar iD, Harris T. Bland iD, David R. Crosslin iD, Siddharth Pratap iD, Ayorinde Cooley iD, Jennifer A. Pacheco iD, Kurt Derek Christensen iD, Emma F. Perez iD, Carrie L. Blout Zawatsky iD, Leora Witkowski iD, Hana Zouk iD, Chunhua Weng iD, Kathleen A. Leppig iD, Patrick M. Abou‐Sleiman iD, Hakon H. Hakonarson iD, Marc S. Williams iD, Yuan Luo iD, Gail Pairitz Jarvik iD, Robert C. Green iD, Wendy K. Chung iD, , Joshua C. Denny iD

Abstract

Knowledge of the clinical spectrum of rare genetic disorders helps in disease management and variant pathogenicity interpretation. Leveraging electronic health record (EHR)-linked genetic testing data from the eMERGE network, we determined the associations between a set of 23 hereditary cancer genes and 3017 phenotypes in 23544 individuals. This phenome-wide association study replicated 45% (184/406) of known gene-phenotype associations (P = 5.1x10-125). Meta-analysis with an independent EHR-derived cohort of 3242 patients confirmed 14 novel associations with phenotypes in the neoplastic, genitourinary, digestive, congenital, metabolic, mental and neurologic categories. Phenotype risk scores (PheRS) based on weighted aggregations of EHR phenotypes accurately predicted variant pathogenicity for at least 50% of pathogenic variants for 8/23 genes. We generated a catalog of PheRS for 7800 variants, including 5217 variants of uncertain significance, to provide empirical evidence of potential pathogenicity. This study highlights the potential of EHR data in genomic medicine.

PDF DOI