Background: Kidney transplant (KTx) recipients receive life-long immunosuppression, which increases the risk of severe infections. Bloodstream infections (BSI) are common after transplantation and are associated with high mortality and morbidity. Prophylactic antibiotic treatment of all KTx recipients does not provide overall benefit, but a personalized strategy of prophylactic treatment of KTx recipients at high risk of BSI with targeted antibiotics has not been assessed.

Primary aim: To determine if prophylactic pivmecillinam in high-risk KTx recipients decreases the incidence of Enterobacterales BSI in the first 1-6 months post-transplantation.

Secondary aim: To assess if prophylactic pivmecillinam reduces all-cause mortality, hospital admissions, graft loss, changes in the gut and urine resistome and microbiome, and increases quality of life in high-risk KTx recipients.

Design and target group: Multi-center double-blinded randomized controlled trial of 150 KTx recipients at high risk of BSI who will be randomized 1:1 to either pivmecillinam 400 mg once daily or placebo from months 1-6 post-transplantation. KTx recipients will be included from Rigshospitalet, Aarhus University Hospital and Odense University Hospital. 60 participants in each study arm will provide urine and stool samples at randomization and at the end of intervention for metagenomic sequencing of the bacterial microbiome and resistome.

Perspectives: This trial will provide evidence necessary to assess if KTx recipients at high risk of BSI benefit from targeted prophylactic antibiotics and address a critical knowledge gap of how to reduce mortality and morbidity due to BSI after KTx. The study will also serve as proof-of-concept for a personalized approach to infection prevention in other populations at high risk of severe infections.

Results from the study may easily be implemented since there is already a clinical set-up for prevention of viral infections in KTx recipients.

BACKGROUND It is estimated that >500,000 adults in Denmark live with chronic kidney disease. Chronic kidney disease can progress, and the most severe manifestation of chronic kidney disease is kidney failure. Kidney transplantation (KTx) is the preferred treatment for patients with kidney failure due to a longer life expectancy compared to other treatment options, including dialysis

(1). Almost 3500 persons in Denmark live with a transplanted kidney, and 250-300 persons receive a KTx each year. However, KTx entails the risk of allograft rejection, which occurs in over 20% of KTx recipients and may lead to graft failure (2). To counteract the risk of graft rejection, KTx recipients receive life-long immunosuppressive medication, which, in turn, increases the risk of infections (3).

Infections are frequent after KTx and may be severe complications. Thus, the incidence of any infection in the first year after KTx is >80%, and infections are a leading cause of mortality (4,5). Bloodstream infections (BSI) are a particular concern due to high mortality, and BSI are associated with increased risk of graft failure (6,7). The investigators have previously found a cumulative incidence of BSI in KTx recipients of 11.8% in the first year post-transplantation, with a 30-day BSI in KTx recipients are often secondary to urinary tract infections and dominated by bacteria of the Enterobacterales order (6,8,9). Urinary tract infections are the most common infectious complication in KTx recipients, and >20% of KTx recipients have recurrent urinary tract infections (10). The investigators and others have previously found urinary tract infections, higher age, comorbidity, acute rejections, and neutropenia to be risk factors for BSI post-transplantation (7,8). To prevent BSI in KTx recipients, screening for bacteriuria has been suggested. However, a randomized clinical trial that included 112 KTx recipients showed no effect of universal screening for bacteriuria and systematic treatment of asymptomatic bacteriuria on the incidence of pyelonephritis (11). Furthermore, the American Society of Transplantation (AST) Infectious Diseases Community of Practice recommends limiting screening for asymptomatic bacteriuria in KTx recipients to the first month post-transplantation due to a lack of evidence of efficacy (12). In conclusion, universal screening for bacteriuria does not work, and new methods for prevention of BSI in KTx recipients are highly warranted. We will target this knowledge gap by use of a risk-based approach where we identify and treat high-risk individuals.

MAIN AIM AND NOVELTY The investigators will conduct a nationwide multicenter randomized double-blind placebo-controlled trial to test the effect of prophylactic pivmecillinam in high risk KTx recipients (n=150) on the incidence of BSI in the first 6 months post-transplantation.

The high risk KTx recipients will be selected using an already developed prediction model to identify KTx recipients with a BSI incidence of 25% in the first year post-transplantation. This approach has not been investigated before, and the study is highly novel with a high potential to prevent BSI and improve health for KTx recipients.

MAIN HYPOTHESIS The investigators hypothesize that the use of targeted prophylactic antibiotics to high risk KTx recipients will decrease the incidence of BSI post-transplantation.

TRIAL DESIGN The study is a nationwide multicenter randomized double-blind placebo-controlled trial. Participants are randomized individually with a 1:1 allocation ratio stratified by study sites. Clinical characteristics will be obtained from national registries, the Scandiatransplant registry, and patient records and microbiology from the Danish microbiology database MiBa (13).

STUDY SETTING The study is anchored at Rigshospitalet, and it will be conducted at all three centers in Denmark responsible for KTx: Rigshospitalet, Aarhus University Hospital, and Odense University Hospital. Participants will be recruited during follow-up at these centers.

ELIGIBILITY CRITERIA Inclusion criteria: Adult (>18 years) KTx recipients at high risk of BSI in the first year post-KTx. Recipients at high risk of BSI will be defined as recipients belonging to a group with a predicted BSI incidence of 25% in the first year post-transplantation by the prediction model.

Exclusion criteria: Recipients who cannot give informed consent or have contraindications for pivmecillinam treatment, including allergy to beta-lactam antibiotics.

INTERVENTION BSI in the first year post-KTx are mainly caused by bacteria of the Enterobacterales order(6,8,9), and in our center >65% of all BSI are caused by Enterobacterales (FIG.2B, 8). The treatment of choice in Denmark is beta-lactam antibiotics covering Gram-negative bacteria, such as pivmecillinam. Despite increasing problems with extended-spectrum beta-lactamase (ESBL), >80% of Enterobacterales at our center are sensitive to pivmecillinam. In this study, the investigators therefore selected pivmecillinam as the intervention. The duration of the intervention is based on the incidence of BSI in KTx in Rigshospitalet from 2010-2019. The intervention will start one month post-transplantation since many KTx receive prophylactic antibiotics immediately after transplantation, and the pathogen composition is more diverse in the first month, with fewer pivmecillinam-sensitive bacteria (8).

At one month post-transplantation, 150 high risk recipients will be randomized 1:1 to either:

• The intervention group (n=75), who will receive pivmecillinam 400 mg once daily from months 1-6 post-transplantation.
• The control group (n=75), who will receive placebo once daily from months 1-6 post-transplantation.

OUTCOMES

The primary outcome is Enterobacterales BSI during the first 1-6 months post-transplantation. The secondary outcomes comprise the following and are assessed at 6 months post-transplantation, unless otherwise stated:

• Severe outcomes: All-cause mortality, hospital admissions, and graft loss or a decrease of 50% in graft function measured as eGFR.
• Infectious outcomes: Enterobacterales urinary tract infections and incidence of all urinary tract infections and BSI.
• Outcomes related to the use of antibiotics: Incidence of infections with multi-drug resistant organisms or Clostridium difficile (C. diff)
• Serious adverse events in accordance with International Council of Harmonisation-Good Clinical Practice (ICH-GCP) guidelines.
• Patient-related outcomes: Patient-reported quality of life (14).

A concern when using long-term antibiotics is development of resistance. As an explorative outcome outcome, the investigators will determine the impact of pivmecillinam on urinary microbiome and resistome to further assess subclinical alterations in microbiome and resistance.

PARTICIPANTS AND STUDY DESIGN (FIG. 1) Trial eligibility is assessed one month post-transplantation, where high risk recipients are included and randomized.

The study participants will be followed for 5 months after inclusion (until 6 months post-transplantation).

Study participants will have three study visits: At inclusion, after two months and after 5 months.

Urinary cultures with resistance patterns will be collected at all three visits.

In addition:

1. At inclusion, participants will answer a questionnaire regarding quality of life (14). In addition, examination of patient records.
2. At two months, compliance with study medication and potential side effects will be assessed.
3. At five months, all participants will answer questionnaires related to quality of life (14), compliance and side effects.

Examination of patient records for graft function, medication, rejection episodes, microbiology (BSI, urinary tract infections, multi-drug resistant organisms, and C. diff), and all-cause mortality, and hospital admissions will be conducted at baseline and at the 5 months visit.

MICROBIOME AND RESISTOME At baseline and 5 months visits, 5 ml of midstream urine will be collected for assessment of microbiome and resistome. Metagenomic DNA will be extracted using the Qiagen Biostic Bacteremia Kit and shotgun metagenomic sequencing will be performed. Following DNA extraction and Illumina library preparation, sequencing will be carried out on the Illumina NovaSeqXPlus at the Norwegian Sequencing Centre, Oslo, Norway. Metagenomic reads will be processed as described (15). Diversity and functional characterization of gut and urinary microbiomes: The investigators will assess the impact of pivmecillinam on the global taxonomic composition of the urinary microbiome using shotgun metagenomic sequencing. Taxonomic profiling will be conducted with MetaPhlAn, and functional and pathway profiling with HUMAnN3. Microbial alpha diversity will be quantified using four indices: richness, Shannon, Simpson, and Gini. Beta diversity will be evaluated using both weighted and unweighted UniFrac distance metrics as previously described (16).

The urinary resistome will be profiled using complementary read-level and genome-resolved approaches. At the read level, resistome profiling will be performed with ARGs-OAP v3.0 and the Structured Antibiotic Resistance Gene (SARG) database (17), which classifies antibiotic resistance genes (ARG) by type and subtype. ARG abundances will be normalized to 16S rRNA gene abundance.

SAMPLE SIZE AND POWER The investigators have developed a prediction model consisting of age, sex, type of induction immunosuppressive treatment, and urinary tract infections in the first months post-KTx. This model can identify a subgroup of high-risk KTx recipients with an incidence of BSI of 25% in the first year post-KTx compared to a low risk group of 6%. In the intervention period (2-6 months post-KTx), the incidence is 23% in the high-risk group. Prevention with antibiotics is usually highly effective, and based on previous BSI in KTx recipients (8), the investigators expect a 76% reduction of BSI in the first 2-6 months. For comparison, use of Bactrim reduces the occurrence of Pneumocystis jirovecii by 85%. Assuming an incidence of BSI of 23% in high-risk recipients in 2-6 months post-KTx, and an alpha of 0.05 (two-sided), the investigators will be able to detect a decrease of 76% with 61 recipients in each arm (total n=122) with a power (1-beta) of 0.80. To account for dropouts and a one-year patient and graft survival of 98% and 96%, respectively (8,15), the investigators aim to include 75 recipients in each arm (total 150).

STATISTICS The primary outcomes (the incidence of Enterobacterales BSI) will be calculated both as intention-to-treat and per-protocol analyses using Gray's test for differences in cumulative incidences.

In addition, the investigatorswill perform intention-to-treat analyses for the secondary outcomes.

Statistical analyses plans will be specified and made publicly available prior to data analyses.

The Statistical analyses plans will include lists of covariates, assessment of secondary efficacy measures, addition of potential interactions and sub-group analyses. Baseline data will be described by treatment group, and the investigators will consider variation in response to treatment by recruitment site.

Inclusion Criteria:

• Adult (>18 years) KTx recipients at high risk of BSI in the first year post-KTx.

Recipients at high risk of BSI will be defined as recipients belonging to a group with a predicted BSI incidence of 25% in the first year post-transplantation by the prediction model.

Exclusion Criteria:

• Recipients who cannot give informed consent or have contraindications for pivmecillinam treatment, including allergy to beta-lactamase antibiotics.