Retrospective validation study of a machine learning-based software for empirical and organism-targeted antibiotic therapy selection.

Errors in antibiotic prescriptions are frequent, often resulting from the inadequate coverage of the infection-causative microorganism. The efficacy of iAST, a machine-learning-based software offering empirical and organism-targeted antibiotic recommendations, was assessed. The study was conducted in a 12-hospital Spanish institution. After model fine-tuning with 27,531 historical antibiograms, 325 consecutive patients with acute infections were selected for retrospective validation. The primary endpoint was comparing each of the top three of iAST's antibiotic recommendations' success rates (confirmed by antibiogram results) with the antibiotic prescribed by the physicians. Secondary endpoints included examining the same hypothesis within specific study population subgroups and assessing antibiotic stewardship by comparing the percentage of antibiotics recommended that belonged to different World Health Organization AWaRe groups within each arm of the study. All of iAST first three recommendations were non-inferior to doctor prescription in the primary endpoint analysis population as well as the secondary endpoint. The overall success rate of doctors' empirical treatment was 68.93%, while that of the first three iAST options was 91.06% (P < 0.001), 90.63% (P < 0.001), and 91.06% (P < 001), respectively. For organism-targeted therapy, the doctor's overall success rate was 84.16%, and that of the first three ranked iAST options was 97.83% (P < 0.001), 94.09% (P < 0.001), and 91.30% (P < 0.001), respectively. In empirical therapy, compared to physician prescriptions, iAST demonstrated a greater propensity to recommend access antibiotics, fewer watch antibiotics, and higher reserve antibiotics. In organism-targeted therapy, iAST advised a higher utilization of access antibiotics. The present study demonstrates iAST accuracy in predicting antibiotic susceptibility, showcasing its potential to promote effective antibiotic stewardship. CLINICAL TRIALS: This study is registered with ClinicalTrials.gov as NCT06174519.

Linezolid induced thrombocytopenia in critically ill patients: Risk factors and development of a machine learning-based prediction model.

INTRODUCTION: Linezolid is an antimicrobial with broad activity against Gram-positive bacteria. Thrombocytopenia is one of its most common side effects often leading to severe complications. The aim of this study is to identify factors related with development of this condition in critically ill patients and to develop and evaluate a predictive machine learning-based model considering easy-to-obtain clinical variables. METHODS: Data was obtained from the Medical Information Mart for Intensive Care III. Patients who received linezolid for over three days were considered, excluding those under 18 years and/or lacking laboratory data. Thrombocytopenia was considered as a platelet decrease of at least 50% from baseline. RESULTS: Three hundred and twenty patients met inclusion criteria of which 63 developed thrombocytopenia and presented significant greater duration of treatment, aspartate-aminotransferase, bilirubin and international normalized ratio; and lower renal clearance and platelet count at baseline. Thrombocytopenia development was associated with a worse outcome (30 days mortality [OR: 2.77; CI95%: 1.87-5.89; P < .001], 60 days mortality [OR: 3.56; CI95%: 2.18-7.26; P < .001]). Thrombocytopenia was also correlated with higher length of hospital stays (35.56 [20.40-52.99] vs 22.69 [10.05-38.61]; P < .001). Median time until this anomaly was of 23 days (CI95%:19.0-NE). Two multivariate models were performed. Accuracy, sensitivity, specificity and AUROC obtained in the best of them were of 0.75, 0.78, 0.62 and 0.80, respectively. CONCLUSION: Linezolid associated thrombocytopenia entails greater mortality rates and hospital stays. Although the proposed predictive model has to be subsequently validated in a real clinical setting, its application could identify patients at risk and establish screening and surveillance strategies.

High-Level Carbapenem Resistance among OXA-48-Producing Klebsiellapneumoniae with Functional OmpK36 Alterations: Maintenance of Ceftazidime/Avibactam Susceptibility.

The aim of this work was to analyze outer membrane porin-encoding genes (ompK35 and ompK36) in a collection of OXA-48 producing Klebsiella pneumoniae, to assess the effect of porin alterations on the susceptibility to ceftazidime/avibactam, and to describe a screening methodology for phenotypic detection of OXA-48-producing K. pneumoniae with disrupted porins. Antimicrobial susceptibility was tested by Microscan and Etest. The genomes of 81 OXA-48-producing K. pneumoniae were sequenced. MLST, detection of antimicrobial resistance genes, and analysis of ompK35 and ompK36 were performed in silico. Tridimensional structures of the OmpK36 variants were assessed. Receiver operating characteristics curves were built to visualize the performance ability of a disk diffusion assay using carbapenems and cefoxitin to detect OmpK36 functional alterations. A wide variety of OmpK36 alterations were detected in 17 OXA-48-producing K. pneumoniae isolates. All displayed a high-level meropenem resistance (MIC ≥ 8 mg/L), and some belonged to high-risk clones, such as ST15 and ST147. Alterations in ompK35 were also observed, but they did not correlate with high-level meropenem resistance. All isolates were susceptible to ceftazidime/avibactam and porin alterations did not affect the MICs of the latter combination. Cefoxitin together with ertapenem/meropenem low inhibition zone diameters (equal or lower than 16 mm) could strongly suggest alterations affecting OmpK36 in OXA-48-producing K. pneumoniae. OXA-48-producing K. pneumoniae with porin disruptions are a cause of concern; ceftazidime/avibactam showed good in vitro activity against them, so this combination could be positioned as the choice therapy to combat the infections caused by this difficult-to-treat isolates.

USP7 and VCPFAF1 define the SUMO/Ubiquitin landscape at the DNA replication fork.

The AAA+ ATPase VCP regulates the extraction of SUMO and ubiquitin-modified DNA replication factors from chromatin. We have previously described that active DNA synthesis is associated with a SUMO-high/ubiquitin-low environment governed by the deubiquitylase USP7. Here, we unveil a functional cooperation between USP7 and VCP in DNA replication, which is conserved from Caenorhabditis elegans to mammals. The role of VCP in chromatin is defined by its cofactor FAF1, which facilitates the extraction of SUMOylated and ubiquitylated proteins that accumulate after the block of DNA replication in the absence of USP7. The inactivation of USP7 and FAF1 is synthetically lethal both in C. elegans and mammalian cells. In addition, USP7 and VCP inhibitors display synergistic toxicity supporting a functional link between deubiquitylation and extraction of chromatin-bound proteins. Our results suggest that USP7 and VCPFAF1 facilitate DNA replication by controlling the balance of SUMO/Ubiquitin-modified DNA replication factors on chromatin.

Coordinating DNA Replication and Mitosis through Ubiquitin/SUMO and CDK1.

Post-translational modification of the DNA replication machinery by ubiquitin and SUMO plays key roles in the faithful duplication of the genetic information. Among other functions, ubiquitination and SUMOylation serve as signals for the extraction of factors from chromatin by the AAA ATPase VCP. In addition to the regulation of DNA replication initiation and elongation, we now know that ubiquitination mediates the disassembly of the replisome after DNA replication termination, a process that is essential to preserve genomic stability. Here, we review the recent evidence showing how active DNA replication restricts replisome ubiquitination to prevent the premature disassembly of the DNA replication machinery. Ubiquitination also mediates the removal of the replisome to allow DNA repair. Further, we discuss the interplay between ubiquitin-mediated replisome disassembly and the activation of CDK1 that is required to set up the transition from the S phase to mitosis. We propose the existence of a ubiquitin-CDK1 relay, where the disassembly of terminated replisomes increases CDK1 activity that, in turn, favors the ubiquitination and disassembly of more replisomes. This model has important implications for the mechanism of action of cancer therapies that induce the untimely activation of CDK1, thereby triggering premature replisome disassembly and DNA damage.

USP7 limits CDK1 activity throughout the cell cycle.

Chemical inhibitors of the deubiquitinase USP7 are currently being developed as anticancer agents based on their capacity to stabilize P53. Regardless of this activity, USP7 inhibitors also generate DNA damage in a p53-independent manner. However, the mechanism of this genotoxicity and its contribution to the anticancer effects of USP7 inhibitors are still under debate. Here we show that, surprisingly, even if USP7 inhibitors stop DNA replication, they also induce a widespread activation of CDK1 throughout the cell cycle, which leads to DNA damage and is toxic for mammalian cells. In addition, USP7 interacts with the phosphatase PP2A and supports its active localization in the cytoplasm. Accordingly, inhibition of USP7 or PP2A triggers very similar changes of the phosphoproteome, including a widespread increase in the phosphorylation of CDK1 targets. Importantly, the toxicity of USP7 inhibitors is alleviated by lowering CDK1 activity or by chemical activation of PP2A. Our work reveals that USP7 limits CDK1 activity at all cell cycle stages, providing a novel mechanism that explains the toxicity of USP7 inhibitors through untimely activation of CDK1.

Comprehensive Statistical Evaluation of Etest®, UMIC®, MicroScan and Disc Diffusion versus Standard Broth Microdilution: Workflow for an Accurate Detection of Colistin-Resistant and Mcr-Positive E. coli.

Four colistin susceptibility testing methods were compared with the standard broth microdilution (BMD) in a collection of 75 colistin-susceptible and 75 mcr-positive E. coli, including ST131 isolates. Taking BMD as reference, all methods showed similar categorical agreement rates (CA) of circa 90%, and a low number of very major errors (VME) (0% for the MicroScan system and Etest®, 0.7% for UMIC®), except for the disc diffusion assay (breakpoint ≤ 11 mm), which yielded false-susceptible results for 8% of isolates. Of note is the number of mcr-positive isolates (17.3%) categorized as susceptible (≤2 mg/L) by the BMD method, but as resistant by the MicroScan system. ST131 mcr-positive E. coli were identified as colistin-resistant by all MIC-based methods. Our results show that applying the current clinical cut-off (>2 mg/L), many mcr-positive E. coli remain undetected, while applying a threshold of >1 mg/L the sensitivity of detection increases significantly without loss of specificity. We propose two possible workflows, both starting with the MicroScan system, since it is automated and, importantly, it categorized all mcr-positive isolates as colistin-resistant. MicroScan should be followed by either BMD or MIC-based commercial methods for colistin resistance detection; or, alternatively, MicroScan, followed by PCR for the mcr screening.