Transcriptome-based Diagnostic Biomarker for Acute Febrile Illness: a Cross-sectional Observational Study

Acute fever is a common medical emergency worldwide, most often caused by bacterial or viral infections. Early and rapid differential diagnosis of infectious diseases is crucial for improving patient outcomes. While pathogen detection remains the gold standard for diagnosing infections, methods like culture are time-consuming and often lack sensitivity. Additionally, the presence of normal colonizing microorganisms, such as bacteria and viruses in the human body, can lead to false positives in pathogen detection. These limitations often compel physicians to rely on empirical antibacterial therapy based on clinical symptoms, inadvertently contributing to antibiotic overuse and the growing problem of bacterial resistance.

Furthermore, the misuse of antibacterial drugs in patients with non-bacterial infections can lead to complications such as secondary infections with Clostridium difficile, liver dysfunction, kidney damage, cytopenia, and alterations in the body's microbiota. Diagnostic markers based on host inflammatory responses offer an alternative approach to infection diagnosis. However, protein biomarkers like procalcitonin, though widely used in clinical settings, are far from ideal for accurately diagnosing bacterial infections, as they are prone to false positives and negatives.

A promising direction is the analysis of host-pathogen interactions at the transcriptome level, particularly the differential gene expression of the host in response to various pathogens. This area of research has gained significant attention recently. Transcriptomic markers derived from patients' peripheral blood have been successfully utilized in diagnosing and studying the pathogenesis of various infectious diseases.

Despite these advances, studies relying on RNA sequencing or transcriptome chip technology require specialized equipment and bioinformatics expertise, making them expensive and challenging to implement in routine clinical practice. Additionally, the results from transcriptome analysis are not easily validated by reverse transcription polymerase chain reaction(RT-PCR), the "gold standard" for RNA quantification. Therefore, to make transcriptome-based diagnostic markers more clinically applicable, there is a need for real-time technologies, such as PCR, to enhance the accuracy of infection diagnosis and reduce the misuse of antibacterial drugs. Currently, research in this area remains limited.