Cost-Effectiveness of Whole-Genome vs Whole-Exome Sequencing Among Children With Suspected Genetic Disorders

Abstract

Importance: The diagnosis of rare diseases and other genetic conditions can be daunting due to vague or poorly defined clinical features that are not recognized even by experienced clinicians. Next-generation sequencing technologies, such as whole-genome sequencing (WGS) and whole-exome sequencing (WES), have greatly enhanced the diagnosis of genetic diseases by expanding the ability to sequence a large part of the genome, rendering a cost-effectiveness comparison between them necessary.

Objective: To assess the cost-effectiveness of WGS compared with WES and conventional testing in children with suspected genetic disorders.

Design, setting, and participants: In this economic evaluation, a bayesian Markov model was implemented from January 1 to June 30, 2023. The model was developed using data from a cohort of 870 pediatric patients with suspected genetic disorders who were enrolled and underwent testing in the Ospedale Pediatrico Bambino Gesù, Rome, Italy, from January 1, 2015, to December 31, 2022. The robustness of the model was assessed through probabilistic sensitivity analysis and value of information analysis.

Main outcomes and measures: Overall costs, number of definitive diagnoses, and incremental cost-effectiveness ratios per diagnosis were measured. The cost-effectiveness analyses involved 4 comparisons: first-tier WGS with standard of care; first-tier WGS with first-tier WES; first-tier WGS with second-tier WES; and first-tier WGS with second-tier WGS.

Results: The ages of the 870 participants ranged from 0 to 18 years (539 [62%] girls). The results of the analysis suggested that adopting WGS as a first-tier strategy would be cost-effective compared with all other explored options. For all threshold levels above €29 800 (US $32 408) per diagnosis that were tested up to €50 000 (US $54 375) per diagnosis, first-line WGS vs second-line WES strategy (ie, 54.6%) had the highest probability of being cost-effective, followed by first-line vs second-line WGS (ie, 54.3%), first-line WGS vs the standard of care alternative (ie, 53.2%), and first-line WGS vs first-line WES (ie, 51.1%). Based on sensitivity analyses, these estimates remained robust to assumptions and parameter uncertainty.

Conclusions and relevance: The findings of this economic evaluation encourage the development of policy changes at various levels (ie, macro, meso, and micro) of international health systems to ensure an efficient adoption of WGS in clinical practice and its equitable access.

Figure 1.. Markov Model Structure

The Markov model structure represents the path followed by an individual pediatric patient suspected of having a genetic disorder. After entering this model in the symptomatic health state at the time of the patient’s first contact, access and continuity of care begin, and individuals with suspected genetic disorders undergo a genetic or genomic test (either a standard genetic test or next-generation sequencing [NGS], depending on the adopted strategy). From there, patients may receive a definitive diagnosis and potentially a different clinical management, a definitive diagnosis with no different clinical management, or remain without a definite diagnosis. In case of a multistep strategy (ie, second-line whole-genome sequencing or second-line whole-exome sequencing), undiagnosed patients undergo a second test, namely NGS, which can be diagnostic and may or may not be followed by a change in clinical management or remain undiagnosed.

Figure 2.. Cost-Effectiveness Plane From Probabilistic Sensitivity Analysis

The cost-effectiveness plane represents the combined distribution of the incremental expected costs (y-axis) and the incremental expected effectiveness (x-axis) from the investigated alternatives. Incremental means the difference between the 2 options, for example, the difference between whole-genome sequencing (WGS) and standard of care (SOC). The dark blue, light blue, gray, and orange dots are individual simulations coming from probabilistic sensitivity analysis; the red dots represent the incremental cost-effectiveness ratios. The area to the right of the vertical line represents the cost-effective region. WES indicates whole-exome sequencing.

Figure 3.. Population Expected Value of Perfect Information (EVPI) Curve

The figure depicts the population EVPI (y-axis) over a range of willingness-to-pay or acceptability thresholds (x-axis) in the Italian pediatric population aged 0 to 18 years.