Write a PREreview

Mycoplasma pneumoniae pneumonia (MPP) has re-emerged globally since 2023, with macrolide-resistant M. pneumoniae (MRMP) exceeding 80% of isolates in China and the Republic of Korea; because tetracyclines and fluoroquinolones are age-restricted in children, no guideline-endorsed antibiotic class outside Japan combines reliable activity against resistant strains with an unrestricted pediatric safety profile. Macrolide resistance prolongs fever and hospitalization without worsening severity, implicating dysregulated host inflammation in the residual burden. We derived an MPP host disease signature from public pediatric transcriptomes and prioritized approved compounds through four executed evidence layers: pathway enrichment, network proximity, LINCS L1000 signature reversal, and literature triangulation. Re-analysis of the largest MPP blood cohort identified 159 differentially expressed genes, and the age-adjusted MPP-versus-viral model reproduced the published signature (2,976 shared genes, 99.8% concordance). The target map converges on NLRP3-IL-1β-pyroptosis and JAK-STAT axes, with a transcriptome-supported AIM2 signal; doxycycline (z = -3.81) was significantly proximal to the disease module, whereas quercetin (z = -3.03) is discounted for annotation-density artifacts. Corticosteroids anchor Tier 1, and JAK inhibitors are the highest-priority testable hypothesis. This hypothesis-generating framework provides a reproducible template for host-directed repurposing in antimicrobial-resistant pediatric respiratory infections.