PNNL

Abstract

Background. Despite extensive medical evaluation many patients with chronic disease remain without a diagnosis. The Undiagnosed Diseases Network (UDN) was established to embrace the most challenging cases in medicine and to characterize the biology of newly discovered diseases. Methods. The UDN was formed in 2014 as a National Institutes of Health (NIH)-funded network of seven clinical sites, two sequencing cores, and a coordinating center. Later a central biorepository, metabolomics core, and model organisms screening center were added. We undertook a retrospective evaluation of the patients referred since the network formed to evaluate the efficacy of this approach. Results. Between September 16, 2015 and May 23, 2017, 1519 patients were referred to the network (53% female). Six hundred and one cases (40%) were accepted for evaluation (53% female). Of those accepted, 192 (32%) had previously undergone exome sequencing. The most common primary symptom categories of accepted patients were neurological (40%), musculoskeletal (9.7%), immunological (6.8%), gastrointestinal (6.5%), and rheumatological (5.7%). The number of completed evaluations on May 23, 2017, was 365, at which time, 100 patients had received diagnoses. In the intervening period, a further 25 received a diagnosis indicating a network diagnosis rate of 34%. Some diagnoses were made from clinical review alone (10%). Many were informed through sequencing the exomes or genomes of the patient or family (72%). A majority of diagnoses (83%) were deemed medically actionable by the clinical team. Indeed, of the total, 30% led to changes in therapy, 24% led to changes in the diagnostic workup, and 28% allowed variant-specific genetic counseling. Through the combined work of the network, a total of 24 new syndromes were defined. Conclusions. A collaborative network with access to state-of-the-art diagnostic tools can expand the medical record of human disease providing actionable clinical diagnoses to a substantial majority of those patients.