Reduction of Antibiotic Resistance Genes in Intestinal Microbiota of Patients With RecurrentClostridium difficileInfection After Fecal Microbiota Transplantation

To the Editor—We read with great interest the recent report by Millan et al studying antibiotic resistance (ABR) genes in patients' microbiota after successful fecal microbiota transplantation (FMT) treatment of recurrent Clostridium difficile infection (rCDI) [1]. The authors reported reduction of both number and diversity of ABR genes in 20 rCDI patients post-FMT compared with pre-FMT fecal samples [1]. Analysis was conducted by using metagenomic sequencing and DNA microarray targeting 354 ABR genes [1]. In our present study, we determined the presence of ABR genes in fecal samples from 3 donors and 8 patients successfully treated with FMT for rCDI [2]. We used quantitative polymerase chain reaction (qPCR) array for 85 ABR genes (Microbial DNA qPCR Array for Antibiotic Resistance genes, Qiagen) and found 33 genes across all samples. The detected genes belonged to 8 different resistance classes: β-lactam ases and beta-lactam resistance (15 genes), fluoroquinolone (6), macrolide, lincosamide, and streptogramin (5), aminoglycoside (2), tetracycline efflux pumps (2), vancomycin (2) and multidrug efflux pumps (1 gene). Similarly to Millan et al, we observed reduction in both the number and diversity of ABR genes in the patients’ samples taken 1 or 2 months after FMT compared with samples prior to FMT (Figure 1). This was apparent regardless of fewer ABR genes being included in our array than in Millan et al's [1]. Thus, our results confirm the recent findings by Millan et al in an independent group of patients. The numbers of antibiotic resistance (ABR) genes (A) and the number of unique and shared genes (B) in fecal samples of donors (n = 3) and patients (n = 8) before and after fecal microbiota transplantation (pre-FMT and post-FMT, respectively) for recurrent Clostridium difficile infection. Box plot shows the medians and quartiles 1 and 3 in boxes and minimum and maximum numbers of ABR genes in the studied groups in whiskers. The Student t test was used to assess the statistical significance between patients pre- and post-FMT (data normally distributed). FMT is an effective treatment for rCDI, with a >90% recovery rate [2–4]. FMT restores the normal composition and diversity of microbiota [1, 4, 5] to resemble that of the healthy donor and consequently provides colonization resistance against C. difficile. Currently, FMT is considered for other indications due to its efficacy in modifying gut microbiota. We agree with the suggestion by Millan et al of using FMT as a treatment option to eradicate multidrug-resistant bacteria from patients [1]. However, it should be acknowledged that although FMT reduces ABR genes in patients, it can also introduce new resistances into the patients’ microbiota. We observed a possible transfer of resistance via FMT; the vanB resistance gene was found in all 3 donors and in all the patients post-FMT, but not pre-FMT. Transfer of nonpathogenic ABR genes, including nonenterococcal transfer of vanB, may be a common scenario in FMT considering the repertoire of resistance genes in fecal donors and healthy adults. Concerning vanB, the gene is commonly found in different intestinal species such as Clostridium, Ruminococcus, and Eggerthella species and even in food-grade lactobacilli [6, 7], and up to 63% of community adults have been found to be nonenterococcal vanB carriers [8, 9]. Nevertheless, our results call for enhanced donor screening programs with consideration of ABR genes too. The importance of even more thorough donor screening is further emphasized by the recent observation that donors’ strains transferred in FMT persist for months [10]. Thus, alongside the increased use of FMT, the donor screening procedures should be constantly improved. Financial support. This work was supported by the Academy of Finland (grant numbers 2584391 and 283088). Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.