The Utility of the Traditional Medical Genetics Diagnostic Evaluation in the Context of Next-Generation Sequencing for Undiagnosed Genetic Disorders. V. Shashi1, A. McConkie-Rosell1, B. Rosell1, K. Schoch1, K. Vellore1, M. McDonald1, Y.-H. Jiang1, P.-X. Xie2, A. Need2, D. Goldstein2 1) Pediatrics, Division of Medical Genetics, Duke University Medical Center, Durham, NC; 2) Center for Human Genome Variation, Duke Universlty Medical Center.

   The diagnostic process used by most medical geneticists is tiered, with clinical assessment followed by sequential laboratory testing, contingent upon the previous tests being negative. Although there have been advances in diagnostic tests there have been no studies of unselected patients to determine the diagnostic yield of this traditional approach. We lack the data required to permit a meaningful comparison of NGS to the traditional paradigm, to provide a clinically and economically grounded diagnostic algorithm for patients with suspected genetic disorders. The purpose of this study was 1) to determine the diagnosis rate, type of disorders diagnosed, number of clinic evaluations to diagnosis and an approximation of the number/cost of the diagnostic tests performed and 2) to determine the appropriate patients to study with NGS and at what stage in their clinical evaluation. Utilizing a retrospective analysis we studied a cohort of 500 unselected consecutive patients (Jan- Sept 2011) from a general genetics clinics at a tertiary medical center. We included all patients who did not have a diagnosis at the time of their initial evaluation and excluded patients referred with a diagnosis or those followed in one of the genetics specialty clinics. Results: Of the 500 patients, 39 did not have a genetic disorder, 212 of the remaining 461 (46%) were diagnosed with a genetic disorder; 72% were diagnosed on the first clinic visit. Diagnostic tests were: Karyotype 12%, FISH 33%, microarray 20%, targeted single gene 13%, gene panel 20%, and biochemical 1%. Conclusions: Based on our findings we have developed a diagnostic algorithm to help guide the clinician. The most critical component of this algorithm is the clinical evaluation that considers the evolving phenotype, emerging diagnostic studies, and the strengths and limitations of the tests. For the majority of diagnosed patients; this diagnosis was considered and diagnostic testing was performed within the first clinical evaluation. Based on the cost of subsequent tests performed on those not diagnosed at the first evaluation, we estimated that the cost per subsequent successful genetic diagnosis was $25,000. For these patients, NGS may be beneficial. Estimating a 50% success rate for NGS in undiagnosed genetic disorders, its application to cases that remain unidentified after the first clinical visit would make the use of NGS in a general genetics clinic clinically and economically beneficial.

You may contact the first author (during and after the meeting) at