PS 3-10 FROM UNCERTAINTY TO ACTION IN PEDIATRIC PRECISION MEDICINE: GENERATING EVIDENCE OF THE COST-EFFECTIVENESS OF CLINICAL EXOME AND GENOME SEQUENCING IN AUTISM SPECTRUM DISORDER

Tuesday, October 25, 2016
Bayshore Ballroom ABC, Lobby Level (Westin Bayshore Vancouver)
Poster Board # PS 3-10

Kate Tsiplova, MSc1, Richard M. Zur, PhD1, Christian R. Marshall, PhD1, D. James Stavropoulos, PhD1, Sergio Pereira, PhD1, Daniele Merico, PhD1, Edwin J. Young, PhD1, Stephen W. Scherer, PhD1 and Wendy J. Ungar, MSc, PhD2, (1)Toronto, ON, Canada, (2)The Hospital for Sick Children Peter Gilgan Centre for Research and Learning, Toronto, ON, Canada
   Purpose: Next generation sequencing, including whole exome (WES) and whole genome sequencing (WGS), is poised to exert a profound effect on diagnostic practice in child health, with implications for health systems, decision-makers, clinicians, patients and families. Currently, chromosomal microarray (CMA) is a first-tier genetic test in autism spectrum disorder (ASD). WES and WGS offer promise in improving diagnostic specificity but these technologies are costly and there have been no economic evaluations in children. The optimal deployment of WES and WGS will depend on their added value and clinical utility. The objectives were to estimate microcosts of CMA, WES and WGS for ASD and to assess the cost-effectiveness of alternative testing strategies.

    Methods: Patient-level and clinical program costs ($CDN) were estimated over 5 years from an institutional perspective. Costs included labour, equipment, depreciation, supplies, follow-up testing, bioinformatics, interpretation, report writing and overhead. WES estimates were based on Illumina HiSeq 2500 and WGS on Illumina HiSeq2500 and HiSeqX. Parameter uncertainty was tested with probabilistic sensitivity analysis. An incremental cost-consequence analysis examined incremental costs per unit of improvement in diagnostic yield for alternative testing strategies.

    Results: The cost per sample was $1978 (95% CI 1928, 2028) for WES, $3551 (3424, 3675) for WGS on Illuminia HiSeq X and $5840 (5573, 6104) for WGS on Illumina HiSeq 2500, compared to $744 (714, 773) for CMA. Reagent supply costs accounted for the largest proportion of costs. Five-year program costs for 300 annual tests were $2.74 million (2.67, 2.81) and $4.94 million (4.76, 5.11) for WES and WGS, respectively. The cost per additional ASD patient with a positive genetic diagnosis was $30,432 when substituting CMA with CMA+WES and $63,078 when substituting CMA with WGS HiSeq X.

    Conclusions: Although newer platforms result in lower laboratory costs for WES and WGS, test costs remain high. The cost to replace older with newer technology exceeded $30,000 per patient to detect an additional pathogenic variant. Research is required to assess the impact of WES/WGS on the ASD care pathway and determine effects on outcomes. This study provides comprehensive costs for use in future economic evaluations of genome and exome sequencing in ASD and allows for a costing model that can be easily adapted to other populations.