Effect of tebipenem pivoxil hydrobromide on the normal gut microbiota of a healthy adult population in Sweden: a randomised controlled trial

Abstract

Background: Antimicrobials cause perturbations in the composition and diversity of the host microbiome. We aimed to compare gut microbiome perturbations caused by oral tebipenem pivoxil hydrobromide (a novel carbapenem) and by amoxicillin-clavulanic acid (an orally administered β-lactam-β-lactam inhibitor combination widely used in clinical practice).

Methods: We did a phase 1, single-centre, randomised, parallel-group, active-control trial to evaluate the effect of tebipenem pivoxil hydrobromide on the human gut microbiota. Healthy participants aged 18 years or older with no documented illnesses during recruitment were enrolled at Karolinska University Hospital (Stockholm, Sweden). Study participants were stratified by sex and block-randomised in a 1:1 ratio to treatment with either tebipenem pivoxil hydrobromide (600 mg orally every 8 h) or amoxicillin-clavulanic acid (500 mg amoxicillin and 125 mg clavulanic acid orally every 8 h). The study included 10 days of treatment (days 1-10) and four follow-up visits (days 14, 21, 90, and 180). The trial was open-label for clinical investigators and patients, but masked for microbiology investigators. Faecal samples were collected at all visits. Sequencing of 16S rDNA was used to measure the diversity metrics, and quantitative culture to quantify selected taxa. The primary outcomes were changes in the α and β diversity and log count of colony-forming units for selected taxa between samples compared with baseline (day 1), and whether any changes reverted during the follow-up period. The analyses were done in the intention-to-treat population. This study was registered with ClinicalTrials.gov (NCT04376554).

Findings: The study was conducted between Jan 23, 2020, and April 6, 2021. 49 volunteers were screened for eligibility, among whom 30 evaluable participants (14 men and 16 women) were assigned: 15 (50%) to the tebipenem pivoxil hydrobromide group and 15 (50%) to the amoxicillin-clavulanic acid group. Baseline characteristics were similar between groups. Complete follow-up was available for all participants, and all participants except one completed treatment as assigned. The diversity metrics showed significant changes from baseline during the treatment period. Significant decreases in richness were observed on days 4-10 (p≤0·0011) in the amoxicillin-clavulanic acid group and on days 4-14 (p≤0·0019) in the tebipenem pivoxil hydrobromide group. Similarly, evenness was significantly decreased during treatment in the amoxicillin-clavulanic acid group (day 4, p=0·030) and the tebipenem pivoxil hydrobromide group (days 4-10, p<0·0001) compared with baseline. Quantitative cultures showed significant decreases in Enterobacterales (days 4-7, p≤0·0030), Enterococcus spp (days 4-14, p=0·025 to p<0·0001), Bifidobacterium spp (days 2-4, p≤0·026), and Bacteroides spp (days 4-10, p≤0·030) in the tebipenem pivoxil hydrobromide group. Similarly, in amoxicillin-clavulanic acid recipients, significant changes were observed in Enterobacterales (days 4-10, p≤0·048), Bifidobacterium spp (days 2-4, p≤0·013), and Lactobacillus spp (days 2-4, p≤0·020). Samples from the follow-up period were not significantly different from those at baseline in β diversity analysis (PERMANOVA, p>0·99). By the end of the study, no significant change was observed compared with baseline in either group. There were no deaths or severe adverse events.

Interpretation: The impact of tebipenem pivoxil hydrobromide on the gut microbiome was similar to that of amoxicillin-clavulanic acid. The safety of antibiotic use with regard to the microbiome should be given attention, as dysbiosis is associated with health and disease.

Funding: Spero Therapeutics.