The Translation of Next Generation Sequencing into Practice: A Cost Analysis

Richard M. Zur, Christian R. Marshall, D. James Stavropoulos, Sergio Pereira, Daniele Merico, Ted Young, Stephen W. Scherer, Wendy J. Ungar

Objectives: Chromosomal microarray (CMA) has replaced karyotyping for testing for the presence of developmental delay (DD). Whole-exome sequencing (WES) is a novel next generation sequencing technology that offers greater diagnostic yield than CMA. The objective of this study was to evaluate the costs of WES for diagnosing DD.

Methods: A microcosting approach was taken whereby testing and sequencing WES protocols were broken down into a series of tasks from venipuncture to interpretation of results. Labour units for each task were either calculated or obtained from the literature. Labour units were calculated for each step in the bioinformatics protocol. Geneticists estimated interpretation time. Labour costs were obtained by multiplying labour units by salaries obtained from hospital records. Volumes of use and unit prices for large and small equipment, computing and supply (reagent) costs were obtained from lab directors and manufacturers. Base case costs and lower and upper bounds for 300 WES tests for DD were estimated over 10 years and discounted at 3%. The effect of varying the total volume of testing for all indications between 300 to 1,000 on the per sample DD WES cost was assessed. Prices for reagents and other items were assumed to remain stable over 10 years. Large and small equipment costs were depreciated over 10 years. Implementation and training costs were not included.

Results: The per sample costs for the WES diagnosis of 300 DD referrals per year ranged from $1855 (minimum $1716 to maximum $2013) in year 1 to $1350 ($1244 to $1466) in year 10. The total ten-year cost of WES for DD referrals was $4,773,325 ($4,406,583 to $5,179,732) for a lab performing 300 tests per year to $3,959,116 (range $3,592,889, $4,309,876) for a lab performing 1,000 tests per year for all indications. Twenty percent of the cost was from labour, 55% from supplies, 20% from small and large equipment, 0.5% from computation and bioinformatics and 5% from overhead costs.

Conclusion: Clinical practitioners are calling for greater use of WES in the future based on an improved diagnostic yield. However, it is essential to fully delineate the microcosts associated with WES testing. These results can used in formal economic evaluations to help inform adoption decisions for the use WES in the diagnosis of developmental delay.