874. Rapid Prediction of Antibiotic Susceptibility in Blood Stream Infections Using Direct Sequencing from Blood Cultures Coupled with Neighbour-Typing Prediction Algorithms.
Bibliographic record
Abstract
Abstract Background Increasing rates of antimicrobial resistance in Gram-negative (GN) bacteria make empiric treatment challenging. Culture-based techniques form the basis of pathogen and antibiotic susceptibility determination yet are limited by long turn-around-times that can prolong the time a patient remains on inappropriate therapy. Direct sequencing of positive blood cultures coupled with neighbour-typing prediction algorithms could help anticipate antibiotic susceptibility and improve selection of empiric therapy. We sought to develop a rapid metagenomic diagnostic workflow and evaluate its performance for blood stream infections. Methods We developed a metagenomic workflow for pathogen identification and prediction of antibiotic susceptibilities outlined in Figure 1. We performed Nanopore-based metagenomic sequencing on positive blood cultures from critically ill patients admitted to a quaternary care center. Isolates belonging to six common GN blood stream pathogens were subjected to a rapid lineage-based prediction algorithm (RASE) that predicts antibiotic susceptibility against a reference set of local isolates. We calculated test performance of predictions compared to phenotypic susceptibility testing. Impact of prediction on post-test probability was calculated for each agent and in aggregate benchmarked against minimum susceptibility thresholds for empiric treatment of 80% for mild infections and 90% for moderate-severe infections. Results We performed metagenomic sequencing and RASE analysis on twelve unique GN blood culture samples with predictions available in 2.5 hours. Across all organisms and antibiotics tested, sensitivity and specificity were 0.87 (95% CI 0.76-0.94) and 0.56 (95% CI 0.31-0.78) although these values approached one for some agents. Impact of RASE predictions on probability of susceptibility is shown in Figure 2. For all antibiotics combined, baseline susceptibility was 79% but improved to 88% or reduced to 47% when RASE predicted a susceptible or resistant phenotype respectively, an effect echoed by specific agents tested. Probability of susceptibility to specific agents or in aggregate after prediction of a susceptible or resistant phenotype by RASE. This was benchmarked against minimum susceptibility thresholds for empiric treatment of 80% for mild infections and 90% for moderate-severe infections. Conclusion We developed a novel metagenomic diagnostic workflow for rapidly predicting antibiotic susceptibility of common GN blood stream pathogens that could improve early selection of empiric therapy. Disclosures All Authors: No reported disclosures
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".