189 - Assessing the Effect of Gestational Age on Automated Prioritization of Sick Newborns for Whole Genome Sequencing

Sunday, April 30, 2023

3:30 PM – 6:00 PM ET

Poster Number: 189
Publication Number: 189.332

Samuel Zoucha, University of Utah School of Medicine, Salt Lake City, UT, United States; Sabrina Malone Jenkins, University of Utah, Salt Lake City, UT, United States; Bennet Peterson, University of Utah, Salt Lake City, UT, United States; Rachel Palmquist, University of Utah School of Medicine, Salt Lake City, UT, United States

Presenting Author(s)

SZ

Samuel Zoucha, MD (he/him/his)

Fellow
University of Utah School of Medicine
Salt Lake City, Utah, United States

Background: As genetic testing becomes faster, cheaper, and more accurate the question of whom to sequence remains critical. It is reported that genetic disease is more prevalent in patients admitted to the intensive care unit. Our group has developed an entirely automated phenotype-driven pipeline, named Mendelian Phenotype Search Engine (MPSE), for selecting acutely ill infants in neonatal intensive care units (NICU) for whole genome sequencing (WGS). MPSE can meet or exceed diagnostic yields obtained through traditional selection procedures that require manual review of clinical histories by specialists (AUC 0.85-0.86). The diagnostic yields are greater than 50% in the first two quartiles of score ranked patients. Given that prematurity may affect the representation of certain human phenotype ontology (HPO) terms, the extent that premature birth may confound this patient prioritization algorithm is unknown.

Objective: We aim to test MPSE on a cohort of premature infants who received WGS to determine if prematurity affects the performance of this tool.

Design/Methods:
In this retrospective study, we used a cohort of infants with a history of NICU admission who received genome sequencing as part of a rapid whole genome sequencing trial (NeoSeq). Our case group consisted of 50 NeoSeq patients (27 preterm and 23 term) with associated phenotypes and gestational age at birth. Our control group consisted of 955 infants (728 preterm and 227 term) born during the same period and admitted to the University of Utah NICU. All infants were scored using our prioritization algorithm for each day of their NICU stay.

Results: Among NeoSeq cases with suspicion of an underlying genetic condition, gestational age was modestly associated with initial score (beta=1.3), while controls showed no association between gestational age and initial score (beta= 0.25). Among case and control patients that eventually reached the score threshold for sequencing recommendation, term infants reached the threshold earlier than preterm infants (7.4 vs 12.1 days; p = 0.01). Diagnostic WGS was seen in 30.7% of preterm infants who crossed the threshold for sequencing vs 38.1% of term infants with no difference between groups.

Conclusion(s): In this study, a group of neonates who underwent WGS showed significant differences in initial scores and time to reach a critical score threshold based on gestational age at birth. There was no difference in diagnosis between these two groups. This study suggests our prioritization algorithm can be used effectively in preterm infants but this population requires more time to cross the sequencing threshold.