Addition of probenecid to oral β-lactam antibiotics: a systematic review and meta-analysis

Abstract Objectives To explore the literature comparing the pharmacokinetic and clinical outcomes from adding probenecid to oral β-lactams. Methods Medline and EMBASE were searched from inception to December 2021 for all English language studies comparing the addition of probenecid (intervention) with an oral β-lactam [flucloxacillin, penicillin V, amoxicillin (± clavulanate), cefalexin, cefuroxime axetil] alone (comparator). ROBINS-I and ROB-2 tools were used. Data on antibiotic therapy, infection diagnosis, primary and secondary outcomes relating to pharmacokinetics and clinical outcomes, plus adverse events were extracted and reported descriptively. For a subset of studies comparing treatment failure between probenecid and control groups, meta-analysis was performed. Results Overall, 18/295 (6%) screened abstracts were included. Populations, methodology and outcome data were heterogeneous. Common populations included healthy volunteers (9/18; 50%) and those with gonococcal infection (6/18; 33%). Most studies were crossover trials (11/18; 61%) or parallel-arm randomized trials (4/18; 22%). Where pharmacokinetic analyses were performed, addition of probenecid to oral β-lactams increased total AUC (7/7; 100%), Cmax (5/8; 63%) and serum t½ (6/8; 75%). Probenecid improved PTA (2/2; 100%). Meta-analysis of 3105 (2258 intervention, 847 control) patients treated for gonococcal disease demonstrated a relative risk of treatment failure in the random-effects model of 0.33 (95% CI 0.20–0.55; I2 = 7%), favouring probenecid. Conclusions Probenecid-boosted β-lactam therapy is associated with improved outcomes in gonococcal disease. Pharmacokinetic data suggest that probenecid-boosted oral β-lactam therapy may have a broader application, but appropriately powered mechanistic and efficacy studies are required.


Introduction
Probenecid, p-(di-n-propylsulphamyl)-benzoic acid, was developed in 1949 with the purpose of decreasing the renal clearance of penicillin. 1 Its mechanism of action is through competitive inhibition of organic anion transporters, which are responsible for excretion of organic agents, such as penicillin. 2 Reduction in renal clearance of penicillin with probenecid demonstrated significant increases in serum exposure, meaning that lower doses of drug were required for similar pharmacokinetic/pharmacodynamic (PK/PD) target attainment. Probenecid's influence on penicillin clearance became mainly academic in the post-war era as the capability to produce more diverse, cheaper and safer β-lactam antibiotics rapidly expanded. 1 Probenecid remains a recommended adjunct in the management of some sexually transmitted infections to support therapeutic target attainment in compartments, such as CSF in neurosyphilis. 3 However, its potential important and broader role in preserving the effectiveness of β-lactams through the optimization of β-lactam PK and dosing schedules needs to be considered, as well as possible adverse events associated with its use, such as nausea and unfavourable drug-drug interactions.
Globally, the WHO Access, Watch and Reserve (AWaRe) criteria require narrow-spectrum antimicrobials, such as the penicillins, to be available in appropriate type, dose and duration to treat common infections. 4 With increasing drug resistance within common causative organisms, such as in streptococci, new methods to optimize the delivery of Access agents and protect the use of broader Watch and Reserve antimicrobials are required. 4,5 It is not always possible to administer higher doses of an oral antibiotic to achieve an optimal PK/PD profile. In some instances, oral drug absorption or gastrointestinal side effects are associated with high doses and limit escalation of therapy. In other situations, augmented renal clearance may make achieving optimal drug exposure difficult. Some agents are not licensed for use at oral doses that would be required to obtain acceptable PK/PD target attainment. Opportunities to deliver oral narrowspectrum agents in an optimized format may offer an attractive opportunity within local antimicrobial stewardship agendas and support the avoidance of prolonged courses of IV treatment in certain infections. 6,7 We explored current and historical literature that compared the use of probenecid with an oral β-lactam antibiotic versus the β-lactam antibiotic alone, describing its impact on PK, clinical outcomes and reported adverse events. The aim was to describe the current literature in support of this approach and identify gaps in knowledge that can be addressed by future mechanistic and efficacy-based research.

Search criteria
We performed a search of MEDLINE and Embase using the search terms outlined in Table S1, available as Supplementary data at JAC Online. Studies in English reporting direct comparison of probenecid plus an oral β-lactam versus the oral β-lactam alone in human subjects were included. Common oral β-lactam antibiotics used in the UK were selected for inclusion. These were flucloxacillin, penicillin V, amoxicillin, ampicillin, amoxicillin/clavulanate, cefalexin and cefuroxime axetil. Only full-text, original research articles comparing the addition of probenecid with the same oral β-lactam antibiotic were included. Articles were required to describe PK/PD, microbiology or adverse event outcomes to be included. Anything published before December 2021 was included and no prior time limit was set. Studies were excluded if they were not in English, were reviews and letters, compared different antimicrobial agents or routes of delivery, or reported on non-human subjects. This review was registered on the PROSPERO database prior to data extraction (registration number: CRD42021298765).

Study selection
Specific literature review software (Covidence, Australia) was used. Two authors (T.M.R. and R.C.W.) independently reviewed abstracts and full texts against inclusion and exclusion criteria. Articles that met screening and eligibility checks were carried forward for full-text review. References of published literature were also reviewed to identify further full texts for inclusion.

Data extraction
Data were extracted by one researcher (T.M.R.), with cross-checking independently performed by a second author (R.C.W. or M.G.). Data extracted included publication details (authors, journal, year of publication), study details (participants, study design, intervention, control, dosing schedules), primary and secondary outcomes (including PK data and/or clinical outcomes) and reported adverse events/toxicity.

Risk of bias
Risk of bias for individual studies was assessed in line with Cochrane recommendations. For non-randomized studies, the Risk Of Bias in Non-randomized Studies of Interventions (ROBINS-I) assessment tool was used. 8 For randomized studies, the Risk of Bias for randomized studies 2 (RoB 2) tool was used. 9 Risk of bias was assessed by two reviewers (T.M.R. and R.C.W.) independently of each other. Where disagreement in domain scoring occurred, a third reviewer assessed the study and differences were discussed to reach consensus.

Studies reporting β-lactam PK
Despite variable β-lactam choice and dose, methods of β-lactam quantification and methods of data analysis, common observations were present. Of 12 studies reporting the effect of probenecid on β-lactam PK as a primary outcome, 7/12 (58%) described the influence on AUC, 8/12 (67%) on serum t ½ , and 8/12 (67%) on peak observed serum concentration (C max ). Two of 12 studies (17%) reported the use of Monte Carlo simulation to estimate PTA. Addition of probenecid to oral β-lactam antibiotics increased total AUC in 7/7 (100%) studies reporting it. β-Lactam C max was significantly increased in 5/8 (63%) and t ½ in 6/8 (75%) of studies reporting these variables. Both studies assessing PTA (2/2; 100%) demonstrated a significant increase in target attainment with the addition of probenecid to β-lactam therapy.

Studies reporting treatment failure
Of the 6/18 (33%) studies reporting on treatment failure as a primary outcome, 4/6 (67%) were included in a meta-analysis comparing the addition of probenecid to an oral β-lactam antibiotic of the same dose on treatment outcome ( Figure 2). 15,17,21,26 One study (17%) could not be included as different doses of ampicillin were used in the intervention and control arms. 13 A further study (1/6; 17%) could not be included due to different dosing schedules between intervention and control arms. 22 All four included studies reported on the outcome of treating gonococcal disease, with microbiological failure at follow-up used to define treatment failure. Three (75%) were randomized studies and one (25%) was observational in design. They contained seven direct comparisons of addition of probenecid to an oral β-lactam antibiotic of fixed dose on treatment outcome in 3105 (2258 intervention and 847 control) patients. The relative risk of treatment failure in the random-effects model was 0.33 (95% CI 0.20-0.55; I 2 = 7%), favouring the addition of probenecid to oral β-lactam regimens.

Side effects and toxicity
The assessment of side effects/toxicity was reported in 11/18 (61%) studies. Of these, 4/11 (36%) observed side effects, with 7/11 (64%) not reporting any observed adverse events. One randomized study identified a higher rate of reported nausea for 1 g cefuroxime axetil with 1 g probenecid (7/57; 12%) versus 1 g cefuroxime axetil alone (1/52; 2%). 17 Within this study, rates of vomiting and diarrhoea were similar. A further study highlighted an increase in observed reports of nausea and dizziness associated with 1 g probenecid twice a day in patients receiving 7 days of treatment for furunculosis. 28 Unfortunately, the observed rate was not quantified by the authors. Allen and colleagues 11 reported one case of nausea associated with an arm containing 1 g of probenecid twice a day in their study of amoxicillin PK in patients with bronchiectasis. The final study to observe side effects reported six patients with nausea from their entire cohort. The authors do not differentiate between those receiving β-lactam antibiotic alone versus β-lactam antibiotic with probenecid. 15 PK data for probenecid and/or β-lactam antibiotics were not provided or not available in a way that allowed evaluation of the impact of drug exposure on these reported outcomes.

Discussion
This review highlights the current paucity of evidence for the use of probenecid to optimize the delivery of oral β-lactam antibiotics. Current data are heterogeneous, use historical methods of drug quantification, and focus predominantly on the management of gonococcal disease. Current evidence suggests that addition of probenecid to oral β-lactam therapy reduces microbiological treatment failures in gonococcal disease compared with use of single doses of an oral β-lactam antibiotic alone. In addition, the influence of probenecid on oral β-lactam PK leads to potentially favourable drug exposures that may enhance target attainment for other infective aetiologies requiring longer courses of antimicrobial therapy, including S. aureus infection.
β-Lactam antibiotics exhibit time-dependent mechanisms of action. In the late 20th and early 21st centuries, optimal PK/PD targets for β-lactams have been explored and defined. The time the free (unbound) concentration of β-lactam spends above an organism's MIC (fT .MIC ) best describes β-lactam PK/PD. 29 Traditionally, targets of greater than 40%-50% fT .MIC are targeted, with evidence that attainment of this target leads to improved patient outcomes. 29 For some infections, such  [30][31][32] To enhance the efficacy of β-lactam antibiotics, different approaches have been trialled, including prolonged and continuous infusions in patients with variable PK. [33][34][35] The benefit of higher doses of oral penicillin for shorter durations has also been demonstrated in conditions such as streptococcal throat infection. 36 Probenecid's ability to potentially prolong terminal t ½ and increase C max and AUC of both oral and IV agents suggests an alternative option to increasing antimicrobial doses or frequency when optimizing PK/PD targets. Everts and colleagues 23 demonstrated significant increases in the PTA for the treatment of S. aureus using oral flucloxacillin co-administered with probenecid compared with oral flucloxacillin alone in healthy volunteers. These preclinical data are further supported by observational studies reporting favourable outcomes for the management of staphylococcal infections using flucloxacillin with probenecid. 37 Furthermore, Grayson and colleagues 38 demonstrated favourable clinical outcomes with IV cefazolin plus probenecid compared with ceftriaxone for the treatment of moderate-to-severe cellulitis as part of a third-generation cephalosporin-sparing regimen.

Current limitations and future steps
Despite emerging observational data supporting the safety and efficacy of probenecid-boosted oral β-lactam therapy, several mechanistic and efficacy questions remain. Current data are limited by the relatively small sample sizes employed in most studies. No experimental data comparing oral β-lactams with and without probenecid have been reported outside of its use in gonococcal disease. Historical analysis of β-lactam PK often determined total antimicrobial exposure from single drug doses and used old methods of quantification, such as tube dilution methods. These methods were often open to wide variation and make direct comparison between studies challenging. Furthermore, the use of total drug concentration does not allow for the active component (free drug concentration) to be described or understood, meaning that the true impact of probenecid on free antibiotic concentration remains to be defined in many cases. Finally, probenecid is known to interact with a number of common medications seen in multi-morbid patients, including paracetamol, non-steroidal antiinflammatories, antipsychotic medications and immunosuppressants. 39 Consideration of these factors on treatment selection and outcomes is lacking from current data. Future work should focus on characterization of the direct efficacy of addition of probenecid to common oral β-lactam antimicrobial dosing regimens. These studies could include the mechanistic characterization of probenecid's influence on free chemically active drug and include assessment of clearance, plasma protein binding and target site concentration attainment. As well as demonstrating enhanced antimicrobial PK using probenecid, an impact on antimicrobial PD, clinical outcomes and toxicity must be clearly demonstrated. Future work should include the assessment and definition of probenecid PK/PD. With improved opportunities to provide therapeutic drug monitoring of both oral β-lactams and probenecid, 40,41 this will further enhance the clinical acceptability of PK manipulation with probenecid and address concerns surrounding potential toxicity, which has not been reported in studies to date.

Conclusions
Probenecid is associated with improved microbiological cure at follow-up when added to oral β-lactam regimens for the treatment of gonococcal disease. Preclinical and observational data suggest that probenecid-boosted oral β-lactam therapy may have a broader application in the future. To define the potential role of probenecid-boosted oral β-lactam regimens, appropriately powered mechanistic and efficacy-based studies to facilitate direct comparison should be conducted.
Cambridge and the University of Warwick. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the NIHR, the Department of Health & Social Care or the UK Health Security Agency (previously PHE); and (3) Professor Alison H. Holmes is an NIHR Senior Investigator.

Funding
This report is independent research funded by the Centre for Antimicrobial Optimisation (CAMO) at Imperial College London.