The Lancet August 01, 2026

Temocillin versus carbapenems for bacteraemia due to third-generation cephalosporin-resistant Enterobacterales in Spain (ASTARTÉ): a multicentre, phase 3, open-label, non-inferiority, randomised clinical trial

Francesco Cogliati Dezza, Lydia Barrera Pulido, Irene Borreguero Borreguero, et al.

Bottom Line

In patients with bacteremia caused by third-generation cephalosporin-resistant Enterobacterales, targeted therapy with intravenous temocillin was non-inferior to carbapenems for 28-day clinical success, validating its use as a carbapenem-sparing alternative.

Key Findings

1. Clinical success at 28 days (mITT population) was achieved in 74% (120 of 163) of patients assigned to temocillin and 73% (121 of 165) of patients assigned to carbapenems.

2. The absolute difference in clinical success was 0.3% (95% CI, -7.7 to infinity), successfully meeting the pre-specified 10% non-inferiority margin (p=0.017 for non-inferiority).

3. Serious adverse events were comparable between groups, reported in 19% (31 of 163) of patients receiving temocillin and 24% (35 of 165) of patients receiving carbapenems.

Study Design

Design

RCT

Open-Label

Sample

328

Patients

Duration

28 days

Median

Setting

29 hospitals, Spain

Population Adults (≥18 years) with monomicrobial bacteraemia due to third-generation cephalosporin-resistant Enterobacterales (3GCR-E) susceptible to meropenem and temocillin, requiring at least 4 days of active intravenous therapy.

Intervention Intravenous temocillin (2 g every 8 hours).

Comparator Intravenous carbapenems (meropenem 1 g every 8 hours, or ertapenem 1 g/day if appropriate).

Outcome Clinical success (composite of clinical cure, no need to stop or change the study drug due to adverse events or clinical failure, absence of recurrence, and survival by day 28) in the modified intention-to-treat (mITT) population.

Study Limitations

• The open-label study design introduces a risk of performance and detection bias, particularly regarding subjective assessments of clinical cure and the need to switch antibiotics.

• Patients had received potentially effective empiric antibiotics for a median of 2 days prior to randomization, which could dampen differences between treatment arms and bias the findings toward non-inferiority.

• The relatively low overall mortality rate suggests the enrolled cohort was less severely ill than the typical hospitalized population with Gram-negative bacteremia, limiting the generalizability of these findings to critically ill patients.

• Global applicability is restricted by medication access; temocillin is currently licensed in only a few European countries and remains largely unavailable in regions like Asia, the Americas, and Sub-Saharan Africa where ESBL prevalence is the highest.

Clinical Significance

The ASTARTÉ trial robustly demonstrates that temocillin is a safe and efficacious definitive therapy for bacteremia caused by third-generation cephalosporin-resistant Enterobacterales (3GCR-E) in hemodynamically stable patients. By providing a viable, narrow-spectrum alternative to carbapenems, temocillin facilitates essential carbapenem-sparing stewardship strategies, helping mitigate the selective evolutionary pressure that drives the emergence of carbapenem-resistant organisms worldwide.

Historical Context

The global spread of Extended-Spectrum Beta-Lactamases (ESBLs) and other 3GCR-E has led to widespread reliance on carbapenems as the default empiric and targeted therapy. However, this escalating utilization has catalyzed a crisis of carbapenem resistance, prompting infectious disease researchers to seek 'carbapenem-sparing' alternatives. While previous major trials investigating alternatives (such as the MERINO trial evaluating piperacillin-tazobactam) unexpectedly demonstrated inferior clinical outcomes, attention shifted back to 'forgotten antibiotics.' Temocillin, a derivative of ticarcillin introduced in the 1980s, offers a highly specific spectrum: it is exceedingly stable against ESBLs and AmpC beta-lactamases but lacks activity against Gram-positive bacteria, anaerobes, and Pseudomonas aeruginosa. The ASTARTÉ trial provides modern, definitive Phase 3 evidence to justify repurposing this neglected drug for modern stewardship protocols.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

What unique structural feature of temocillin allows it to resist hydrolysis by extended-spectrum beta-lactamases (ESBLs), and why is it preferred over broader-spectrum penicillins in the context of 3GCR Enterobacterales?

Key Response

Temocillin is a 6-alpha-methoxy derivative of ticarcillin. The methoxy group provides steric hindrance, protecting the beta-lactam ring from hydrolysis by ESBLs and AmpC beta-lactamases. This allows it to effectively target 3GCR Enterobacterales while its lack of activity against Gram-positives, anaerobes, and Pseudomonas minimizes collateral microbiome damage compared to broader agents.

Resident

Given the ASTARTÉ trial findings supporting temocillin for targeted therapy, under what specific clinical scenarios would it be inappropriate to use temocillin as empiric monotherapy for a patient with suspected 3GCR Enterobacterales bacteremia?

Key Response

Temocillin lacks intrinsic activity against Pseudomonas aeruginosa, anaerobes, and Gram-positive organisms (like MRSA or Enterococcus). Therefore, it should not be used as empiric monotherapy in situations where these organisms are likely, such as intra-abdominal infections without adequate source control, severe neutropenic fever, or suspected polymicrobial sepsis, emphasizing its role primarily as targeted therapy once susceptibilities are known.

Fellow

Temocillin is a time-dependent antibiotic often administered via continuous infusion in critically ill patients. How do the PK/PD properties of temocillin justify this dosing strategy, and how might augmented renal clearance (ARC) in early sepsis affect the non-inferiority margin observed in the ASTARTÉ trial?

Key Response

As a beta-lactam, temocillin's efficacy depends on the time the free drug concentration exceeds the MIC (fT>MIC). Continuous infusion maximizes this parameter, which is vital for high-inoculum infections like bacteremia. Fellows must recognize that ARC in early sepsis can lead to rapid clearance and subtherapeutic levels, which could bias the trial toward inferiority if appropriate therapeutic drug monitoring or optimized continuous dosing protocols were not utilized in the experimental arm.

Attending

While the ASTARTÉ trial validates temocillin as a carbapenem-sparing agent, how should antimicrobial stewardship programs integrate this finding without inadvertently selecting for novel resistance mechanisms, and what are the practical barriers to implementation?

Key Response

Attendings must balance individual patient care with macro-level stewardship. Widespread use of temocillin spares carbapenems, reducing selective pressure for carbapenem-resistant Enterobacterales (CRE), but risks selecting for OXA-48 or metallo-beta-lactamase producers which hydrolyze temocillin. Practical barriers include the need to update local laboratory testing panels for temocillin susceptibility and educating clinicians to avoid using it empirically where Pseudomonas is a risk.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

The ASTARTÉ trial utilized an open-label, non-inferiority design. How does the open-label nature of this study introduce the risk of ascertainment bias regarding the 28-day clinical success outcome, and how might this bias statistically push the results toward the null, thereby artificially inflating the claim of non-inferiority?

Key Response

In an open-label non-inferiority trial, knowledge of treatment allocation can influence subjective components of 'clinical success' (e.g., decisions to escalate therapy, adjunctive care). If investigators subtly standardize care or expect similar outcomes, they may unconsciously rate both arms similarly (ascertainment bias). This biases the treatment difference toward zero, increasing the likelihood that the lower bound of the confidence interval falls within the non-inferiority margin, potentially resulting in a false-positive claim of non-inferiority.

Journal Editor

As a peer reviewer evaluating the ASTARTÉ manuscript, what critical threats to external validity would you flag regarding the molecular epidemiology of the ESBLs in the Spanish cohort compared to a global population?

Key Response

A rigorous reviewer would highlight that Spain's regional resistance epidemiology may heavily feature CTX-M ESBL enzymes, which temocillin resists well. However, in regions with higher baseline rates of OXA-48-like carbapenemases (which confer resistance to temocillin but might test susceptible to 3rd-generation cephalosporins in vitro if poorly expressed) or specific AmpC variants, temocillin might fail. The manuscript must explicitly detail the molecular characterization of the isolates to ensure the results aren't merely a reflection of favorable local microbial ecology.

Guideline Committee

Current IDSA and ESCMID guidelines strongly recommend carbapenems as first-line therapy for ESBL-producing Enterobacterales bacteremia, explicitly warning against piperacillin-tazobactam based on the MERINO trial. Does the ASTARTÉ trial provide sufficient evidence to introduce temocillin as a preferred carbapenem-sparing alternative in the next guideline iteration?

Key Response

Yes, the ASTARTÉ trial provides crucial Phase 3 RCT data demonstrating non-inferiority to carbapenems, unlike the MERINO trial where piperacillin-tazobactam failed to show non-inferiority for ESBL bacteremia. Guidelines should be updated to include temocillin as a strong recommendation (Level of Evidence I) for targeted step-down therapy for 3GCR Enterobacterales bacteremia, provided the isolate is susceptible, offering a much-needed, evidence-based carbapenem-sparing strategy.

Clinical Landscape

Noteworthy Related Trials

2018

MERINO Trial

n = 379 · JAMA

Tested

Piperacillin-tazobactam

Population

Patients with ceftriaxone-resistant E. coli or K. pneumoniae bloodstream infection

Comparator

Meropenem

Endpoint

30-day all-cause mortality

Key result: Piperacillin-tazobactam did not meet non-inferiority criteria compared to meropenem for 30-day mortality.

2018

AIDA Trial

n = 406 · Lancet Infect Dis

Tested

Colistin plus meropenem

Population

Patients with severe carbapenem-resistant Gram-negative infections

Comparator

Colistin monotherapy

Endpoint

Clinical failure at 14 days

Key result: Combination therapy of colistin and meropenem was not superior to colistin alone in preventing clinical failure.

2020

CREDIBLE-CR Trial

n = 152 · Lancet Infect Dis

Tested

Cefiderocol

Population

Patients with severe carbapenem-resistant Gram-negative infections

Comparator

Best available therapy

Endpoint

Clinical cure and all-cause mortality

Key result: Cefiderocol demonstrated similar clinical efficacy to best available therapy for highly resistant Gram-negative infections.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis