P-1306. Priority Bacterial Pathogens Co-Resistance Networks Reveal Predictable Novel Pathways for Pan-Drug Resistance

TL;DR

This study maps antibiotic resistance pathways in priority bacteria to predict how resistance spreads and leads to pan-drug resistance.

Contribution

The study identifies predictable resistance pathways and shared gateway resistances leading to pan-drug resistance in priority bacterial pathogens.

Findings

Resistance pathways for K. pneumoniae, E. coli, and A. baumannii show interconnected beta-lactam and aminoglycoside co-resistance.

Early fluoroquinolone resistance acts as a gateway to multidrug resistance across species.

Predictable resistance progression patterns were found, enabling early identification of high-risk pathways.

Abstract

The rising global concern for the development of pan-drug resistant (PDR) infections is alarming. We mapped the existing structural high antimicrobial co-resistance networks for potential novel pathways for resistance trajectories that may lead to PDR across WHO priority bacterial pathogens.Figure 1.Co-resistance network analysis (% Resistance > 10) for the WHO bacterial priority pathogens.Figure 2.Heat map showing co-resistance network matrices for each WHO priority bacterial pathogen in the network analysis

Co-resistance network analysis (% Resistance > 10) for the WHO bacterial priority pathogens.

Heat map showing co-resistance network matrices for each WHO priority bacterial pathogen in the network analysis

We analysed 151,358 clinical isolates data collected over 14 years (2010–2023) from different hospitals in the UAE with their AMR profiles. Pathogen-drug co-resistance networks were constructed using antibiotic pairs with ≥10% resistance. Nodes were annotated by drug class and WHO AWaRe classification. Network structures were quantified by nodes, edges, density, clustering and degree.Figure 3.Resistance pathways derived from the co-resistance network matrices for drug pairs with min (%Resistance 1, %Resistance 2) ≥ 10%, retained as weighted edges showing pathways to MDR for each pathogen.Figure 4.Universal novel resistance pathways highlighting key resistance trajectories towards MDR and PDR.

Resistance pathways derived from the co-resistance network matrices for drug pairs with min (%Resistance 1, %Resistance 2) ≥ 10%, retained as weighted edges showing pathways to MDR for each pathogen.

Universal novel resistance pathways highlighting key resistance trajectories towards MDR and PDR.

The resistance trajectories revealed species-specific antibiotic resistance pathways ending to MDR/PDR. The co-resistance network-based analysis captured statistically inferred transitions between drugs co-occurring ≥10%. K. pneumoniae, E. coli, and A. baumannii exhibited highly interconnected networks with extensive beta-lactam and aminoglycoside co-resistance. PDR trajectory pathways for each organism typically passed through MDR transition nodes before converging on carbapenem, vancomycin, or linezolid resistance. Transition modelling confirmed early fluoroquinolone resistance as a shared gateway to MDR, followed by organism-specific paths culminating in PDR. There were consistent resistance progression patterns that offered universal complementary frameworks for anticipating resistance evolution and optimizing empiric therapy.

Drug resistance evolution is predictable and convergent, not random. Despite species differences, resistance followed a convergent topology anchored on early fluoroquinolone resistance and structured progression to PDR. This framework enables early identification of high-risk resistance pathways and supports precision antimicrobial stewardship

All Authors: No reported disclosures

Figures