Transcriptome assists prognosis of disease severity in respiratory syncytial virus infected infants

Area of Science:

• Virology
• Genomics
• Pediatric Medicine

Background:

• Respiratory syncytial virus (RSV) infections present a wide spectrum of illness, from mild colds to severe lower respiratory tract infections.
• Predicting disease progression in infants upon initial presentation is clinically challenging, as RSV can rapidly escalate to critical illness.

Purpose of the Study:

• To investigate the existence of a genomic signature capable of accurately predicting the clinical course and severity of respiratory syncytial virus infections in hospitalized infants.
• To develop a potential prognostic tool to support early clinical decision-making for RSV patients.

Main Methods:

• Utilized early blood microarray transcriptome profiles from 39 hospitalized infants, retrospectively assessing disease severity.
• Applied support vector machine learning to age- and sex-standardized transcriptomic data to identify a predictive gene signature.
• Validated the identified signature on an independent cohort of 53 infants.

Main Results:

• An 84-gene signature was identified that effectively discriminated between infants with less severe and those with the most severe RSV infections.
• The gene signature achieved an area under the receiver operating characteristic curve (AUC) of 0.966 (cross-validation) and 0.858 (independent validation).
• Combining the gene signature with infant age and sex improved prediction accuracy to an AUC of 0.971.

Conclusions:

• The identified 84-gene signature demonstrates high accuracy in predicting RSV disease severity in hospitalized infants.
• This genomic signature holds promise as a basis for developing a novel prognostic test to enhance clinical management strategies for RSV.
• Integrating this signature with clinical factors like age and sex offers a robust approach for predicting RSV outcomes.