A Machine Learning Decision Support Tool Optimizes Whole Genome Sequencing Utilization in a Neonatal Intensive Care Unit.

The Mendelian Phenotype Search Engine (MPSE), a clinical decision support tool using Natural Language Processing and Machine Learning, helped neonatologists expedite decisions to whole genome sequencing (WGS) to diagnose patients in the Neonatal Intensive Care Unit. After the MPSE was introduced, utilization of WGS increased, time to ordering WGS decreased, and WGS diagnostic yield increased.

Rapid Whole-Genome Sequencing and Clinical Management in the PICU: A Multicenter Cohort, 2016-2023.

OBJECTIVES: Analysis of the clinical utility of rapid whole-genome sequencing (rWGS) outside of the neonatal period is lacking. We describe the use of rWGS in PICU and cardiovascular ICU (CICU) patients across four institutions. DESIGN: Ambidirectional multisite cohort study. SETTING: Four tertiary children's hospitals. PATIENTS: Children 0-18 years old in the PICU or CICU who underwent rWGS analysis, from May 2016 to June 2023. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: A total of 133 patients underwent clinical, phenotype-driven rWGS analysis, 36 prospectively. A molecular diagnosis was identified in 79 patients (59%). Median (interquartile range [IQR]) age was 6 months (IQR 1.2 mo-4.6 yr). Median time for return of preliminary results was 3 days (IQR 2-4). In 79 patients with a molecular diagnosis, there was a change in ICU management in 19 patients (24%); and some change in clinical management in 63 patients (80%). Nondiagnosis changed management in 5 of 54 patients (9%). The clinical specialty ordering rWGS did not affect diagnostic rate. Factors associated with greater odds ratio (OR [95% CI]; OR [95% CI]) of diagnosis included dysmorphic features (OR 10.9 [95% CI, 1.8-105]) and congenital heart disease (OR 4.2 [95% CI, 1.3-16.8]). Variables associated with greater odds of changes in management included obtaining a genetic diagnosis (OR 16.6 [95% CI, 5.5-62]) and a shorter time to genetic result (OR 0.8 [95% CI, 0.76-0.9]). Surveys of pediatric intensivists indicated that rWGS-enhanced clinical prognostication (p < 0.0001) and contributed to a decision to consult palliative care (p < 0.02). CONCLUSIONS: In this 2016-2023 multiple-PICU/CICU cohort, we have shown that timely genetic diagnosis is feasible across institutions. Application of rWGS had a 59% (95% CI, 51-67%) rate of diagnostic yield and was associated with changes in critical care management and long-term patient management.

Genomic sequencing has a high diagnostic yield in children with congenital anomalies of the heart and urinary system.

Background: Congenital heart defects (CHD) and congenital anomalies of the kidney and urinary tract (CAKUT) account for significant morbidity and mortality in childhood. Dozens of monogenic causes of anomalies in each organ system have been identified. However, even though 30% of CHD patients also have a CAKUT and both organs arise from the lateral mesoderm, there is sparse overlap of the genes implicated in the congenital anomalies for these organ systems. We sought to determine whether patients with both CAKUT and CHD have a monogenic etiology, with the long-term goal of guiding future diagnostic work up and improving outcomes. Methods: Retrospective review of electronic medical records (EMR), identifying patients admitted to Rady Children's Hospital between January 2015 and July 2020 with both CAKUT and CHD who underwent either whole exome sequencing (WES) or whole genome sequencing (WGS). Data collected included demographics, presenting phenotype, genetic results, and mother's pregnancy history. WGS data was reanalyzed with a specific focus on the CAKUT and CHD phenotype. Genetic results were reviewed to identify causative, candidate, and novel genes for the CAKUT and CHD phenotype. Associated additional structural malformations were identified and categorized. Results: Thirty-two patients were identified. Eight patients had causative variants for the CAKUT/CHD phenotype, three patients had candidate variants, and three patients had potential novel variants. Five patients had variants in genes not associated with the CAKUT/CHD phenotype, and 13 patients had no variant identified. Of these, eight patients were identified as having possible alternative causes for their CHD/CAKUT phenotype. Eighty-eight percent of all CAKUT/CHD patients had at least one additional organ system with a structural malformation. Conclusions: Overall, our study demonstrated a high rate of monogenic etiologies in hospitalized patients with both CHD and CAKUT, with a diagnostic rate of 44%. Thus, physicians should have a high suspicion for genetic disease in this population. Together, these data provide valuable information on how to approach acutely ill patients with CAKUT and CHD, including guiding diagnostic work up for associated phenotypes, as well as novel insights into the genetics of CAKUT and CHD overlap syndromes in hospitalized children.

Approaches to long-read sequencing in a clinical setting to improve diagnostic rate.

Over the past decade, advances in genetic testing, particularly the advent of next-generation sequencing, have led to a paradigm shift in the diagnosis of molecular diseases and disorders. Despite our present collective ability to interrogate more than 90% of the human genome, portions of the genome have eluded us, resulting in stagnation of diagnostic yield with existing methodologies. Here we show how application of a new technology, long-read sequencing, has the potential to improve molecular diagnostic rates. Whole genome sequencing by long reads was able to cover 98% of next-generation sequencing dead zones, which are areas of the genome that are not interpretable by conventional industry-standard short-read sequencing. Through the ability of long-read sequencing to unambiguously call variants in these regions, we discovered an immunodeficiency due to a variant in IKBKG in a subject who had previously received a negative genome sequencing result. Additionally, we demonstrate the ability of long-read sequencing to detect small variants on par with short-read sequencing, its superior performance in identifying structural variants, and thirdly, its capacity to determine genomic methylation defects in native DNA. Though the latter technical abilities have been demonstrated, we demonstrate the clinical application of this technology to successfully identify multiple types of variants using a single test.

Better and faster is cheaper.

The rapid pace of advancement in genomic sequencing technology has recently reached a new milestone, with a record-setting time to molecular diagnosis of a mere 8 h. The catalyst behind this achievement is the accumulation of evidence indicating that quicker results more often make an impact on patient care and lead to healthcare cost savings. Herein, we review the diagnostic and clinical utility of rapid whole genome and rapid whole exome sequencing, the associated reduction in healthcare costs, and the relationship between these outcome measures and time-to-diagnosis.

Cost Efficacy of Rapid Whole Genome Sequencing in the Pediatric Intensive Care Unit.

The diagnostic and clinical utility of rapid whole genome sequencing (rWGS) for critically ill children in the intensive care unit (ICU) has been substantiated by multiple studies, but comprehensive cost-effectiveness evaluation of rWGS in the ICU outside of the neonatal age group is lacking. In this study, we examined cost data retrospectively for a cohort of 38 children in a regional pediatric ICU (PICU) who received rWGS. We identified seven of 17 patients who received molecular diagnoses by rWGS and had resultant changes in clinical management with sufficient clarity to permit cost and quality adjusted life years (QALY) modeling. Cost of PICU care was estimated to be reduced by $184,846 and a total of 12.1 QALYs were gained among these seven patients. The total cost of rWGS for patients and families for the entire cohort (38 probands) was $239,400. Thus, the net cost of rWGS was $54,554, representing $4,509 per QALY gained. This quantitative, retrospective examination of healthcare utilization associated with rWGS-informed medicine interventions in the PICU revealed approximately one-third of a QALY gained per patient tested at a cost per QALY that was approximately one-tenth of that typically sought for cost-effective new medical interventions. This evidence suggests that performance of rWGS as a first-tier test in selected PICU children with diseases of unknown etiology is associated with acceptable cost-per-QALY gained.