Pharmacogenomic (PGx) guidelines provide recommendations that can improve drug efficacy and reduce toxicity by tailoring therapy to genetic variation. Their application in pediatric intensive care units (PICU) has not been widely characterized. This study aimed to determine how often medications with established PGx guidelines are administered in a tertiary PICU and to evaluate whether diagnostic exome sequencing can capture PGx-relevant phenotypes for these medications.
This was a retrospective cohort study integrating electronic medical record (EMR) data with exome sequencing results. The work was conducted at Morgan Stanley Children’s Hospital of NewYork-Presbyterian, a single-center tertiary care children’s hospital. The analysis combined two data sources covering different timeframes: EMR medication data from 2020–2024 and research exome sequencing performed between 2015 and 2023.
Two analytic cohorts were reported. The first cohort consisted of 4,939 children admitted to the PICU between 2020 and 2024; medication administration records from this population were used to quantify exposure to drugs with established PGx guidelines. The second cohort comprised 192 children admitted to the PICU who underwent exome sequencing for research purposes from 2015 to 2023; this group was analyzed to determine the proportion of patients in whom exome data could identify PGx phenotypes such as metabolizer status.
Among the 4,939 PICU patients, 37.2% (n=1,837) received at least one medication that has an established pharmacogenomic guideline. A subset, 14.4% (n=712), received two or more medications with such guidelines. Across the exposed patients, medications implicated PGx effects related to variants in 20 different genes. The most frequently implicated gene was CYP2C9, involved in 17.3% (n=853) of patients who received guideline-linked medications.
The authors estimated that 8.2% of the total PICU population studied received medications for which applying PGx-guided recommendations would have altered clinical management. This figure represents the proportion of patients whose treatment choices could potentially have been modified by available guideline-based genetic recommendations.
In the cohort of 192 PICU patients with research exome sequencing, at least one metabolizer phenotype relevant to drug response was identifiable in 62% (n=119) of individuals. This finding indicates that exome sequencing can detect a substantial fraction of clinically relevant PGx phenotypes, although the study also assessed and discussed the limitations of exome data for comprehensive PGx determination.
The report notes that exome sequencing has both strengths and weaknesses for uncovering PGx phenotypes. While many metabolizer phenotypes were detectable, the methodology and exome capture design may not identify all variant types relevant to PGx guidance (for example, copy-number variants, structural variants, or noncoding variants) and coverage may vary by gene. The source text presents overall detection rates but does not provide a gene-level breakdown beyond the most commonly implicated gene in the medication exposure analysis.
The integrated EMR–exome analysis demonstrates that a substantial proportion of critically ill children in this tertiary PICU received medications for which pharmacogenomic guidelines exist. With 37.2% of admissions exposed to at least one such medication and an estimated 8.2% potentially requiring a management change under PGx guidance, the findings highlight a tangible opportunity to incorporate pharmacogenomic information into critical care prescribing to support more personalized therapy.
The exome sequencing data suggest that many PGx-relevant metabolizer phenotypes can be identified from research-grade exomes, supporting feasibility of genomic-informed prescribing in the PICU. However, the authors emphasize that exome-based detection is not exhaustive and that implementation would need to consider which genes and variant types are clinically actionable and reliably captured by sequencing platforms.
Strengths of the study include linkage of real-world EMR medication exposure with genomic data and evaluation of both population-level exposure frequencies and individual-level genotype-derived phenotypes. The single-center design enabled detailed integration of institutional records and research exome data.
Limitations reported or inherent to the described design include the retrospective nature of the cohorts and the single-center setting, which may limit generalizability. The preprint status indicates the work has not undergone peer review. Specific limitations of exome sequencing for PGx identification are acknowledged: not all clinically relevant variant types may be detected by exome sequencing, and capture or coverage variability across genes can affect phenotype ascertainment. The source did not provide a complete gene-by-gene sensitivity analysis or a full list of the 20 implicated genes beyond reporting CYP2C9 as the most common.
Ethical oversight was obtained: the Columbia University Medical Center institutional review board approved the research (protocol AAAV0962). The authors declared no competing interests.
Many critically ill children admitted to this tertiary PICU received medications with established pharmacogenomic guidelines, and a meaningful minority received medications for which PGx-guided recommendations would have changed management. Exome sequencing identified at least one metabolizer phenotype in 62% of sequenced patients, demonstrating the potential for genomic data to inform prescribing in the PICU while also underscoring the technical and coverage limitations of exome-based PGx detection. The findings point to an actionable opportunity to explore integration of PGx into pediatric critical care, with further work needed on implementation strategies, comprehensive genotype detection, and assessment of clinical outcomes.