Machine Learning-Led Optimization of Combination Therapy: Confronting the Public Health Threat of Extensively Drug Resistant Gram-Negative Bacteria.

Developing optimized regimens for combination antibiotic therapy is challenging and often performed empirically over many clinical studies. Novel implementation of a hybrid machine-learning pharmacokinetic/pharmacodynamic/toxicodynamic (ML-PK/PD/TD) approach optimizes combination therapy using human PK/TD data along with in vitro PD data. This study utilized human population PK (PopPK) of aztreonam, ceftazidime/avibactam, and polymyxin B along with in vitro PDs from the Hollow Fiber Infection Model (HFIM) to derive optimal multi-drug regimens de novo through implementation of a genetic algorithm (GA). The mechanism-based PD model was constructed based on 7-day HFIM experiments across 4 clinical, extensively drug resistant Klebsiella pneumoniae isolates. GA-led optimization was performed using 13 different fitness functions to compare the effects of different efficacy (60%, 70%, 80%, or 90% of simulated subjects achieving bacterial counts of 102 CFU/mL) and toxicity (66% of simulated subjects having a target polymyxin B area under the concentration-time curve [AUC] of 100 mg·h/L and aztreonam AUC of 1,332 mg·h/L) on the optimized regimen. All regimens, except those most heavily weighted for toxicity prevention, were able to achieve the target efficacy threshold (102 CFU/mL). Overall, GA-based regimen optimization using preclinical data from animal-sparing in vitro studies and human PopPK produced clinically relevant dosage regimens similar to those developed empirically over many years for all three antibiotics. Taken together, these data provide significant insight into new therapeutic approaches incorporating ML to regimen design and treatment of resistant bacterial infections.

Mechanistic Insights to Combating NDM- and CTX-M-Coproducing Klebsiella pneumoniae by Targeting Cell Wall Synthesis and Outer Membrane Integrity.

Metallo-ß-lactamase (MBL)-producing Gram-negative bacteria cause infections associated with high rates of morbidity and mortality. Currently, a leading regimen to treat infections caused by MBL-producing bacteria is aztreonam combined with ceftazidime-avibactam. The purpose of the present study was to evaluate and rationally optimize the combination of aztreonam and ceftazidime-avibactam with and without polymyxin B against a clinical Klebsiella pneumoniae isolate producing NDM-1 and CTX-M by use of the hollow fiber infection model (HFIM). A novel de-escalation approach to polymyxin B dosing was also explored, whereby a standard 0-h loading dose was followed by maintenance doses that were 50% of the typical clinical regimen. In the HFIM, the addition of polymyxin B to aztreonam plus ceftazidime-avibactam significantly improved bacterial killing, leading to eradication, including for the novel de-escalation dosing strategy. Serial samples from the growth control and monotherapies were explored in a Galleria mellonella virulence model to assess virulence changes. Weibull regression showed that low-level ceftazidime resistance and treatment with monotherapy resulted in increased G. mellonella mortality (P < 0.05). A neutropenic rabbit pneumonia model demonstrated that aztreonam plus ceftazidime-avibactam with or without polymyxin B resulted in similar bacterial killing, and these combination therapies were statistically significantly better than monotherapies (P < 0.05). However, only the polymyxin B-containing combination therapy produced a statistically significant decrease in lung weights (P < 0.05), indicating a decreased inflammatory process. Altogether, adding polymyxin B to the combination of aztreonam plus ceftazidime-avibactam for NDM- and CTX-M-producing K. pneumoniae improved bacterial killing effects, reduced lung inflammation, suppressed resistance amplification, and limited virulence changes.

Determining the optimal dosing of a novel combination regimen of ceftazidime/avibactam with aztreonam against NDM-1-producing Enterobacteriaceae using a hollow-fibre infection model.

BACKGROUND: MBL-producing strains of Enterobacteriaceae are a major public health concern. We sought to define optimal combination regimens of ceftazidime/avibactam with aztreonam in a hollow-fibre infection model (HFIM) of MBL-producing strains of Escherichia coli and Klebsiella pneumoniae. METHODS: E. coli ARLG-1013 (blaNDM-1, blaCTX-M, blaCMY, blaTEM) and K. pneumoniae ARLG-1002 (blaNDM-1, blaCTXM-15, blaDHA, blaSHV, blaTEM) were studied in the HFIM using simulated human dosing regimens of ceftazidime/avibactam and aztreonam. Experiments were designed to evaluate the effect of staggered versus simultaneous administration, infusion duration and aztreonam daily dose (6 g/day versus 8 g/day) on bacterial killing and resistance suppression. Prospective validation experiments for the most active combination regimens were performed in triplicate to ensure reproducibility. RESULTS: Staggered administration of the combination (ceftazidime/avibactam followed by aztreonam) was found to be inferior to simultaneous administration. Longer infusion durations (2 h and continuous infusion) also resulted in enhanced bacterial killing relative to 30 min infusions. The rate of killing was more pronounced with 8 g/day versus 6 g/day aztreonam combination regimens for both tested strains. In the prospective validation experiments, ceftazidime/avibactam with aztreonam dosed every 8 and 6 h, respectively (ceftazidime/avibactam 2/0.5 g every 8 h + aztreonam 2 g every 6 h), or ceftazidime/avibactam with aztreonam as continuous infusions resulted in maximal bacterial killing and resistance suppression over 7 days. CONCLUSIONS: Simultaneous administration of aztreonam 8 g/day given as a continuous or 2 h infusion with ceftazidime/avibactam resulted in complete bacterial eradication and resistance suppression. Further study of this combination is needed with additional MBL-producing Gram-negative pathogens. The safety of this double ß-lactam strategy also warrants further study in Phase 1 clinical trials.

Gene expression studies for the analysis of domoic acid production in the marine diatom Pseudo-nitzschia multiseries.

BACKGROUND: Pseudo-nitzschia multiseries Hasle (Hasle) (Ps-n) is distinctive among the ecologically important marine diatoms because it produces the neurotoxin domoic acid. Although the biology of Ps-n has been investigated intensely, the characterization of the genes and biochemical pathways leading to domoic acid biosynthesis has been limited. To identify transcripts whose levels correlate with domoic acid production, we analyzed Ps-n under conditions of high and low domoic acid production by cDNA microarray technology and reverse-transcription quantitative PCR (RT-qPCR) methods. Our goals included identifying and validating robust reference genes for Ps-n RNA expression analysis under these conditions. RESULTS: Through microarray analysis of exponential- and stationary-phase cultures with low and high domoic acid production, respectively, we identified candidate reference genes whose transcripts did not vary across conditions. We tested eleven potential reference genes for stability using RT-qPCR and GeNorm analyses. Our results indicated that transcripts encoding JmjC, dynein, and histone H3 proteins were the most suitable for normalization of expression data under conditions of silicon-limitation, in late-exponential through stationary phase. The microarray studies identified a number of genes that were up- and down-regulated under toxin-producing conditions. RT-qPCR analysis, using the validated controls, confirmed the up-regulation of transcripts predicted to encode a cycloisomerase, an SLC6 transporter, phosphoenolpyruvate carboxykinase, glutamate dehydrogenase, a small heat shock protein, and an aldo-keto reductase, as well as the down-regulation of a transcript encoding a fucoxanthin-chlorophyll a-c binding protein, under these conditions. CONCLUSION: Our results provide a strong basis for further studies of RNA expression levels in Ps-n, which will contribute to our understanding of genes involved in the production and release of domoic acid, an important neurotoxin that affects human health as well as ecosystem function.