Infectious Disease · Medical Literature
Benzylpenicillin versus flucloxacillin or cloxacillin for the treatment of penicillin-susceptible Staphylococcus aureus bacteraemia (SNAP): an international, multicentre, open-label, non-inferiority randomised controlled trial.
Abstract
Previously considered rare, penicillin-susceptible Staphylococcus aureus (PSSA) bacteraemia has re-emerged worldwide. Although benzylpenicillin might offer advantages in terms of pharmacokinetic and adverse effect profiles, anti-staphylococcal penicillins are recommended for serious infections because of concern over undetected penicillin resistance. We aimed to compare benzylpenicillin with anti-staphylococcal penicillins (cloxacillin or flucloxacillin) for the treatment of PSSA bacteraemia in adults. This investigator-initiated, international, multicentre, open-label, non-inferiority, randomised controlled trial was conducted within the PSSA silo of the backbone domain of the ongoing S aureus Network Adaptive Platform (SNAP) trial. We enrolled patients of all ages who were admitted with S aureus bacteraemia at 67 hospitals across Australia, New Zealand, Canada, Israel, the Netherlands, the UK, Singapore, and South Africa. Herein, we report results for adult patients (aged ≥18 years). Participants were randomly allocated 1:1 to receive either benzylpenicillin, or flucloxacillin or cloxacillin. Trial staff, staff caring for participants, and participants were aware of the treatment allocated and received. Cloxacillin was used only where flucloxacillin was not available (ie, Canada, Israel, Singapore, and South Africa). Recommended standard dosing for benzylpenicillin was 1·8 g intravenously once every 4 h or 2·4 g once every 6 h; for flucloxacillin 2·0 g intravenously once every 6 h; and for cloxacillin 2·0 g intravenously once every 4 h. The primary outcome was all-cause mortality 90 days after platform entry, assessed in the intention-to-treat population, including all patients with available data. The primary outcome was analysed by use of a hierarchical Bayesian logistic regression model, with non-inferiority of benzylpenicillin defined as an adjusted odds ratio (OR) of less than 1·20. Analyses occurred after every 500 participants reached 90-day follow-up. The SNAP trial is registered with ClinicalTrials.gov (NCT05137119) and is ongoing. Following the fourth interim analysis, the data and safety monitoring committee recommended ceasing recruitment before prespecified stopping thresholds were reached because of increased acute kidney injury (AKI) in participants receiving flucloxacillin or cloxacillin, and the PSSA silo was closed for recruitment on Aug 7, 2024. Between Feb 18, 2022, and June 21, 2024, of 493 adults with PSSA bacteraemia, 125 were randomly assigned to the flucloxacillin or cloxacillin group and 156 to the benzylpenicillin group. The median age was 67 years (IQR 56-77), 87 (31%) were female, and 194 (69%) were male. 21 (14%) of 152 in the benzylpenicillin group and 26 (22%) of 121 in the flucloxacillin or cloxacillin group met the primary outcome of death at 90 days (adjusted OR 0·67, 95% credible interval [CrI] 0·35-1·28), with a posterior probability of benzylpenicillin non-inferiority of 96·1% and of benzylpenicillin superiority of 88·9%. AKI occurred in 17 (11%) of 153 patients in the benzylpenicillin group and 27 (22%) of 124 patients in the flucloxacillin or cloxacillin group (adjusted OR 0·50, 95% CrI 0·26-0·94), corresponding to a posterior probability of non-inferiority of benzylpenicillin of 99·8% and superiority of benzylpenicillin of 98·4%. Seven total serious adverse reactions were reported from six (4%) of 156 participants in the benzylpenicillin group and ten were reported from nine (7%) of 125 participants in the flucloxacillin or cloxacillin group. Although the prespecified non-inferiority criterion for benzylpenicillin was not met, the probability of benzylpenicillin being non-inferior for mortality, along with a reduction in risk of AKI, indicates that benzylpenicillin should be preferred over flucloxacillin or cloxacillin for treatment of PSSA bacteraemia in adults. National Health and Medical Research Council, Medical Research Future Fund, Canadian Institutes of Health Research, Accelerating Clinical Trials Consortium, UMC Utrecht, ZonMW Good Use of Medicines programme, Health Research Council of New Zealand, Starship Foundation, National Healthcare Group Fund, National Medical Research Council, National Institute for Health and Care Research, Medical Research Council, and Paterson Family Foundation.
Clinical relevance
In adults with laboratory-confirmed penicillin-susceptible Staphylococcus aureus (PSSA) bacteraemia, is benzylpenicillin non-inferior to anti-staphylococcal penicillins (flucloxacillin or cloxacillin) with respect to 90-day all-cause mortality? __SNAPSHOT__ {"studyDesign":"Investigator-initiated, international, multicentre, open-label, non-inferiority randomised controlled trial (Bayesian adaptive platform design; PSSA silo of the SNAP backbone domain)","setting":"67 hospitals across Australia, New Zealand, Canada, Israel, the Netherlands, the United Kingdom, Singapore, and South Africa","sampleSize":"n=281 adults randomised (benzylpenicillin: n=156; flucloxacillin or cloxacillin: n=125); 493 adults with PSSA bacteraemia were enrolled in the broader silo, with 281 contributing to the primary efficacy analysis","population":"Adults aged ≥18 years admitted with laboratory-confirmed PSSA bacteraemia (penicillin susceptibility confirmed by local testing); PSSA re-emerging internationally and estimated to represent up to 25% of all clinical S. aureus isolates in some regions","intervention":"Benzylpenicillin 1.8 g (3 million units) IV every 4 hours or 2.4 g (4 million units) IV every 6 hours (at treating clinician's discretion)","comparator":"Flucloxacillin 2 g IV every 6 hours or cloxacillin (dose not reported in available abstract/sources); allocated 1:1","primaryOutcome":"All-cause mortality at 90 days, analysed by hierarchical Bayesian logistic regression; non-inferiority defined as an adjusted OR <1.20 (corresponding to an absolute mortality difference of approximately 2.5% assuming 15% control-arm mortality)","followUp":"90 days","fundingSource":"Investigator-initiated; supported by regional government and non-profit funders across participating countries (specific funding agencies not reported in available sources)"}
Key findings
In the primary efficacy analysis, 90-day all-cause mortality occurred in 21/152 (14%) participants in the benzylpenicillin group versus 26/121 (22%) in the flucloxacillin or cloxacillin group (adjusted OR 0.67, 95% CI 0.35–1.28). The posterior probability that benzylpenicillin was non-inferior to anti-staphylococcal penicillins was 96.1%, and the posterior probability of superiority was 88.9%. The prespecified Bayesian non-inferiority threshold required a posterior probability >99%, which was NOT met. The absolute difference in 90-day mortality was 8 percentage points (22% vs. 14%), favouring benzylpenicillin; NNT to prevent one death = approximately 13 (calculated), though this observation must be interpreted cautiously given early termination and the wide confidence interval crossing 1.0. Acute kidney injury (AKI) occurred in 11% of patients in the benzylpenicillin group versus 22% in the flucloxacillin or cloxacillin group; this safety signal—specifically AKI excess in the comparator arm—prompted the Data and Safety Monitoring Committee to recommend early closure of the PSSA silo after the fourth interim analysis (August 7, 2024), before prespecified stopping thresholds for efficacy or non-inferiority were reached. The majority of AKI events were mild (stage 1 in 7% benzylpenicillin vs. 15% flucloxacillin/cloxacillin groups), though precise breakdown across all AKI stages was not fully reported in available sources.
Practice implications
The SNAP PSSA trial established that benzylpenicillin has a high but formally insufficient posterior probability (96.1% vs. required >99%) of being non-inferior to anti-staphylococcal penicillins for 90-day mortality in adults with confirmed PSSA bacteraemia. The clinical conclusion therefore rests on a near-threshold non-inferiority probability combined with a consistent and clinically meaningful reduction in AKI, a major adverse effect of flucloxacillin and cloxacillin. Directional consistency across mortality, AKI, intention-to-treat, and sensitivity analyses strengthens the interpretation that benzylpenicillin is at least as effective and is safer. Findings apply to adults with rigorously confirmed PSSA at institutions capable of accurate penicillin susceptibility testing; they should not be extrapolated to settings lacking reliable blaZ detection, to methicillin-susceptible S. aureus (MSSA) that is penicillin-resistant, or to patients with high endocarditis burden (subgroup was small). The key unanswered question—whether benzylpenicillin should be preferred over cefazolin for PSSA bacteraemia—was not addressed by this trial. Because early stopping inflates uncertainty and the formal non-inferiority criterion was not crossed, some caution in immediate universal practice change is warranted, though the totality of evidence supports preferring benzylpenicillin over anti-staphylococcal penicillins where reliable susceptibility testing is available. Only the abstract and secondary reports were available for this summary; exact subgroup data, protocol-adherent analyses, and sensitivity analyses were not independently verifiable. __LIMITATIONS__ Early termination for safety (excess AKI in the comparator arm) before the prespecified non-inferiority efficacy threshold was crossed; early stopping reduces sample size and precision and may exaggerate treatment effects in both directions.,The formal Bayesian non-inferiority criterion (posterior probability >99%) was not met (observed: 96.1%), meaning the primary hypothesis was technically not proven by the trial's own operating characteristics.,Open-label design: lack of blinding may introduce performance bias in clinical management decisions and ascertainment of soft endpoints such as AKI.,Relatively small sample size (n=281 analysable in primary analysis) for a mortality endpoint in a disease with 15–25% 90-day mortality; the confidence interval around the OR (0.35–1.28) is wide and includes the null.,Generalisability is contingent on access to high-quality penicillin susceptibility testing (combined automated MIC screening and confirmatory phenotypic testing); results do not apply to settings where blaZ-positive strains may be misclassified as susceptible.,The direct comparison of benzylpenicillin versus cefazolin—now the most clinically relevant head-to-head question given concurrent SNAP cefazolin data—was not evaluated.,Endocarditis and other deep-focus infection subgroups were small, limiting applicability to patients with complicated bacteraemia.,Country- and centre-level variation in susceptibility testing methodology, antibiotic dosing practices, and adjunctive care may limit homogeneity of the treatment effect across sites.,Full publication not independently reviewed; this summary is based on the published abstract and secondary reports. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In adults with laboratory-confirmed PSSA bacteraemia, benzylpenicillin demonstrated a near-threshold but formally unmet Bayesian non-inferiority probability for 90-day mortality (96.1% vs. required >99%, adjusted OR 0.67 [95% CI 0.35–1.28]) alongside a clinically important reduction in AKI compared with flucloxacillin or cloxacillin. The totality of evidence—directionally consistent across mortality, AKI, and sensitivity analyses—supports preferring benzylpenicillin over anti-staphylococcal penicillins for PSSA bacteraemia in adults where reliable penicillin susceptibility testing is available. The critical remaining question of benzylpenicillin versus cefazolin for PSSA was not addressed and requires further study before a definitive first-line agent can be established.