Antimicrobial resistance of bacterial pathogens is an increasing clinical problem and alternative approaches to antibiotic chemotherapy are needed. One of these approaches is the use of lytic bacterial viruses known as phage therapy. We aimed to assess the efficacy of phage therapy in preclinical animal models of bacterial infection. In this systematic review and meta-analysis, MEDLINE/Ovid, Embase/Ovid, CINAHL/EbscoHOST, Web of Science/Wiley, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, and Google Scholar were searched from inception to Sept 30, 2021. Studies assessing phage efficacy in animal models were included. Only studies that assessed the efficacy of phage therapy in treating established bacterial infections in terms of survival and bacterial abundance or density were included. Studies reporting only in-vitro or ex-vivo results and those with incomplete information were excluded. Risk-of-bias assessment was performed using the Systematic Review Centre for Laboratory Animal Experimentation tool. The main endpoints were animal survival and tissue bacterial burden, which were reported using pooled odds ratios (ORs) and mean differences with random-effects models. The I² measure and its 95% CI were also calculated. This study is registered with PROSPERO, CRD42022311309. Of the 5084 references screened, 124 studies fulfilled the selection criteria. Risk of bias was high for 70 (56%) of the 124 included studies; therefore, only studies classified as having a low-to-moderate risk of bias were considered for quantitative data synthesis (n=32). Phage therapy was associated with significantly improved survival at 24 h in systemic infection models (OR 0·08 [95% CI 0·03 to 0·20]; I²=55% [95% CI 8 to 77]), skin infection (OR 0·08 [0·04 to 0·19]; I² = 0% [0 to 79]), and pneumonia models (OR 0·13 [0·06 to 0·31]; I²=0% [0 to 68]) when compared with placebo. Animals with skin infections (mean difference -2·66 [95% CI -3·17 to -2·16]; I² = 95% [90 to 96]) and those with pneumonia (mean difference -3·35 [-6·00 to -0·69]; I² = 99% [98 to 99]) treated with phage therapy had significantly lower tissue bacterial loads at 5 ± 2 days of follow-up compared with placebo. Phage therapy significantly improved animal survival and reduced organ bacterial loads compared with placebo in preclinical animal models. However, high heterogeneity was observed in some comparisons. More evidence is needed to identify the factors influencing phage therapy performance to improve future clinical application. Swiss National Foundation and Swiss Heart Foundation.

Antibiotic resistance and its rapid dissemination around the world threaten the efficacy of currently-used medical treatments and call for novel, innovative approaches to manage multi-drug resistant infections. Phage therapy, i.e., the use of viruses (phages) to specifically infect and kill bacteria during their life cycle, is one of the most promising alternatives to antibiotics. It is based on the correct matching between a target pathogenic bacteria and the therapeutic phage. Nevertheless, correctly matching them is a major challenge. Currently, there is no systematic method to efficiently predict whether phage-bacterium interactions exist and these pairs must be empirically tested in laboratory. Herein, we present our approach for developing a computational model able to predict whether a given phage-bacterium pair can interact based on their genome. Based on public data from GenBank and phagesDB.org, we collected more than a thousand positive phage-bacterium interactions with their complete genomes. In addition, we generated putative negative (i.e., non-interacting) pairs. We extracted, from the collected genomes, a set of informative features based on the distribution of predictive protein-protein interactions and on their primary structure (e.g. amino-acid frequency, molecular weight and chemical composition of each protein). With these features, we generated multiple candidate datasets to train our algorithms. On this base, we built predictive models exhibiting predictive performance of around 90% in terms of F1-score, sensitivity, specificity, and accuracy, obtained on the test set with 10-fold cross-validation. These promising results reinforce the hypothesis that machine learning techniques may produce highly-predictive models accelerating the search of interacting phage-bacteria pairs.

The optimal method for delivering phages in the context of ventilator-associated pneumonia (VAP) is unknown. In the current study, we assessed the utility of aerosolized phages (aerophages) for experimental methicillin-resistant Staphylococcus aureus (MRSA) pneumonia. Rats were ventilated for 4 hours before induction of pneumonia. Animals received one of the following: (1) aerophages; (2) intravenous (IV) phages; (3) a combination of IV and aerophages; (4) IV linezolid; or (5) a combination of IV linezolid and aerophages. Phages were administered at 2, 12, 24, 48, and 72 hours, and linezolid was administered at 2, 12, 24, 36, 48, 60, and 72 hours. The primary outcome was survival at 96 hours. Secondary outcomes were bacterial and phage counts in tissues and histopathological scoring of the lungs. Aerophages and IV phages each rescued 50% of animals from severe MRSA pneumonia (P < .01 compared with placebo controls). The combination of aerophages and IV phages rescued 91% of animals, which was higher than either monotherapy (P < .05). Standard-of-care antibiotic linezolid rescued 38% of animals. However, linezolid and aerophages did not synergize in this setting (55% survival). Aerosolized phage therapy showed potential for the treatment of MRSA pneumonia in an experimental animal model and warrants further investigation for application in humans.

Background: Accurate biomarkers to diagnose infection are lacking. Studies reported good performance of pancreatic stone protein (PSP) to detect infection. The objective of the study was to determine the performance of PSP in diagnosing infection across hospitalized patients and calculate a threshold value for that purpose. Methods: A systematic search across Cochrane Central Register of Controlled Trials and MEDLINE databases (1966–March 2019) for studies on PSP published in English using ‘pancreatic stone protein’, ‘PSP’, ‘regenerative protein’, ‘lithostatin’ combined with ‘infection’ and ‘sepsis’ found 44 records. The search was restricted to the five trials that evaluated PSP for the initial detection of infection in hospitalized adults. Individual patient data were obtained from the investigators of all eligible trials. Data quality and validity was assessed according to PRISMA guidelines. We choose a fixed-effect model to calculate the PSP cut-off value that best discriminates infected from non-infected patients. Results: Infection was confirmed in 371 of 631 patients. The median (IQR) PSP value of infected versus uninfected patients was 81.5 (30.0–237.5) versus 19.2 (12.6–33.57) ng/ml, compared to 150 (82.70–229.55) versus 58.25 (15.85–120) mg/l for C-reactive protein (CRP) and 0.9 (0.29–4.4) versus 0.15 (0.08–0.5) ng/ml for procalcitonin (PCT). Using a PSP cut-off of 44.18 ng/ml, the ROC AUC to detect infection was 0.81 (0.78–0.85) with a sensitivity of 0.66 (0.61–0.71), specificity of 0.83 (0.78–0.88), PPV of 0.85 (0.81–0.89) and NPV of 0.63 (0.58–0.68). When a model combining PSP and CRP was used, the ROC AUC improved to 0.90 (0.87–0.92) with higher sensitivity 0.81 (0.77–0.85) and specificity 0.84 (0.79–0.90) for discriminating infection from non-infection. Adding PCT did not improve the performance further. Conclusions: PSP is a promising biomarker to diagnose infections in hospitalized patients. Using a cut-off value of 44.18 ng/ml, PSP performs better than CRP or PCT across the considered studies. The combination of PSP with CRP further enhances its accuracy.

As pharmacokinetics after burn trauma are difficult to predict, we conducted a 3-year prospective, monocentric, randomized, controlled trial to determine the extent of under- and overdosing of antibiotics and further evaluate the impact of systematic therapeutic drug monitoring (TDM) with same-day real-time dose adaptation to reach and maintain antibiotic concentrations within the therapeutic range. Forty-five consecutive burn patients treated with antibiotics were prospectively screened. Forty fulfilled the inclusion criteria; after one patient refused to participate and one withdrew consent, 19 were randomly assigned to an intervention group (patients with real-time antibiotic concentration determination and subsequent adaptations) and 19 were randomly assigned to a standard-of-care group (patients with antibiotic administration at the physician's discretion without real-time TDM). Seventy-three infection episodes were analyzed. Before the intervention, only 46/82 (56%) initial trough concentrations fell within the range. There was no difference between groups in the initial trough concentrations (adjusted hazard ratio = 1.39 [95% confidence interval {CI}, 0.81 to 2.39], P = 0.227) or the time to reach the target. However, thanks to real-time dose adjustments, the trough concentrations of the intervention group remained more within the predefined range (57/77 [74.0%] versus 48/85 [56.5%]; adjusted odd ratio [OR] = 2.34 [95% CI, 1.17 to 4.81], P = 0.018), more days were spent within the target range (193 days/297 days on antibiotics [65.0%] versus 171 days/311 days in antibiotics [55.0%]; adjusted OR = 1.64 [95% CI, 1.16 to 2.32], P = 0.005), and fewer results were below the target trough concentrations (25/118 [21.2%] versus 44/126 [34.9%]; adjusted OR = 0.47 [95% CI, 0.26 to 0.87], P = 0.015). No difference in infection outcomes was observed between the study groups. Systematic TDM with same-day real-time dose adaptation was effective in reaching and maintaining therapeutic antibiotic concentrations in infected burn patients, which prevented both over- and underdosing. A larger multicentric study is needed to further evaluate the impact of this strategy on infection outcomes and the emergence of antibiotic resistance during long-term burn treatment. (This study was registered with the ClinicalTrials.gov platform under registration no. NCT01965340 on 27 September 2013.)

Background: The potential of phage therapy for the treatment of endovascular Staphylococcus aureus infections remains to be evaluated. Methods and Results: The efficacy of a phage cocktail combining Herelleviridae phage vB_SauH_2002 and Podoviriae phage 66 was evaluated against a methicillin‐sensitive S. aureus strain in vitro and in vivo in a rodent model of experimental endocarditis. Six hours after bacterial challenge, animals were treated with (1) the phage cocktail. (2) subtherapeutic flucloxacillin dosage, (3) combination of the phage cocktail and flucloxacillin, or (4) saline. Bacterial loads in cardiac vegetations at 30 hours were the primary outcome. Secondary outcomes were phage loads at 30 hours in cardiac vegetations, blood, spleen, liver, and kidneys. We evaluated phage resistance 30 hours post infection in vegetations of rats under combination treatment. In vitro, phages synergized against S. aureus planktonic cells and the cocktail synergized with flucloxacillin to eradicated biofilms. In infected animals, the phage cocktail achieved bacteriostatic effect. The addition of low‐dose flucloxacillin elevated bacterial suppression (∆ of −5.25 log ₁₀ colony forming unit/g [CFU/g] versus treatment onset, P <0.0001) and synergism was confirmed (∆ of −2.15 log ₁₀ CFU/g versus low‐dose flucloxacillin alone, P <0.01). Importantly, 9/12 rats given the combination treatment had sterile vegetations at 30 hours. In vivo phage replication was partially suppressed by the antibiotic and selection of resistance to the Podoviridae component of the phage cocktail occurred. Plasma‐mediated inhibition of phage killing activity was observed in vitro. Conclusions: Combining phages with a low‐dose standard of care antibiotic represents a promising strategy for the treatment of S. aureus infective endocarditis.