WO2023156628A1

WO2023156628A1 - Methods for predicting severity of dysbiosis caused by treatment with an antibiotic

Info

Publication number: WO2023156628A1
Application number: PCT/EP2023/054100
Authority: WO
Inventors: Surbhi MALHOTRA-KUMAR; Matilda BERKELL; Herman Goossens; Mohamed Mysara AHMED
Original assignee: Universiteit Antwerpen; Studiecentrum Voor Kernenergie / Centre D'etude De L'energie Nucléaire
Priority date: 2022-02-18
Filing date: 2023-02-17
Publication date: 2023-08-24

Abstract

The invention concerns a method for predicting severity of dysbiosis after treatment of a subject with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject. The invention further relates to method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject. The invention further provides related kits of parts.

Description

METHODS FOR PREDICTING SEVERITY OF DYSBIOSIS CAUSED BY TREATMENT WITH AN ANTIBIOTIC

FIELD OF THE INVENTION

The invention is broadly in the field of medicine, more precisely in the area of diagnostics and therapeutics. In particular, the invention concerns methods for predicting severity of dysbiosis after treatment of a subject with an antibiotic or for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic. The invention also relates to methods for selecting a particular antibiotic or class of antibiotics for treatment of a subject.

BACKGROUND OF THE INVENTION

The human intestinal microbiota constitutes a complex community of organisms, predominantly bacteria, closely linked to human health. As a result of antibiotic treatment, disturbances in microbial diversity and composition are introduced that disrupt essential microbial metabolic reactions, mucosal and epithelial barrier integrity, and enteropathogenic colonization resistance. A single orally administered course of antibiotics can result in distinct perturbations in both bacterial load and diversity in both the intestinal and respiratory flora, and such perturbations can be further enhanced by prolonged treatment, dosage, and non-antibiotic drug interactions. Consequently, up to 35% of patients treated with antibiotics develop antibiotic-associated diarrhea (AAD), a condition characterized by microbial disruption and is associated with a considerable economic burden that can further result in severe conditions like septicemia and colitis-related deaths.

However, antibiotic compound-specific perturbations remain largely uncharacterized. Previous studies aiming at determining antibiotic-specific perturbations in the intestinal microbiota were primarily limited to small study cohorts of healthy individuals within narrow geographical regions.

In view thereof, there remains a need in the art for methods predictive of compound-specific antibiotic-induced dysbiosis in patients and/or for methods which could be used to guide antibiotic treatment to reduce the frequency of dysbiosis-associated adverse events like AAD.

SUMMARY OF THE INVENTION

The present inventors have found by extensive experimental testing that samples of patients after antibiotic treatment clustered into two distinct microbial community types, one characterised by low to moderate dysbiosis (MCT1) and the other by severe dysbiosis (MCT2) as indicated by enrichment of potential enteropathogens like enterococci and Escherichia/Shigella spp. The latter group was further characterized by ultra-low microbial diversity and developed antibiotic- associated diarrhea (AAD) at higher rates and higher frequency than patients who displayed MCT1 microbial communities after treatment. When comparing taxonomical differences in microbial composition between MCT1 and MCT2 communities, it was found that prior to antibiotic treatment MCT1 communities were characterized by elevated levels of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium spp. In contrast, these taxa were largely absent in MCT2 communities prior to antibiotic treatment.

Accordingly, a first aspect of the invention relates to a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in a fecal sample or gut microbiota sample obtained from the subject. In the methods as taught herein, the fecal sample or gut microbiota sample may be obtained from the subject prior to or during antibiotic treatment, preferably prior to antibiotic treatment.

Preferably, the invention provides a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject. In the methods as taught herein, the fecal sample or gut microbiota sample may be obtained from the subject prior to or during antibiotic treatment, preferably prior to antibiotic treatment.

Hence, an aspect of the invention provides a method for predicting prior to an envisaged treatment of a subject with an antibiotic whether the subject is at risk of developing severe dysbiosis after treatment with the antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject. In the methods as taught herein, the fecal sample or gut microbiota sample may be obtained from the subject prior to or during antibiotic treatment, preferably prior to antibiotic treatment.

In embodiments of the methods as taught herein, the method comprises: (a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject; (b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and where applicable one or more of Akkermansia,

Porphyromonas, or Bifidobacterium bacteria, preferably wherein the reference value represents a reference subject having (developed) severe dysbiosis after treatment with an antibiotic; and (c) predicting that the subject is at risk of developing severe dysbiosis after treatment with an antibiotic if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is the same or decreased compared with the reference value. Preferably, the reference value represents the quantity of said bacteria in a sample of a reference subject having developed severe dysbiosis after treatment with an antibiotic. The present method advantageously allows to determine prior to antibiotic treatment whether a subject is at risk of developing severe dysbiosis after antibiotic treatment. In other words, the present method allows to predict microbial community development of severe dysbiosis before treatment with an antibiotic. The present method further allows to determine whether a subject is at risk of developing severe dysbiosis after treatment with a particular class of antibiotics or a particular antibiotic. Hence, in embodiments of the methods as taught herein, the method is for predicting whether a subject is at risk of developing severe dysbiosis after treatment with a particular antibiotic or a particular class of antibiotics.

A further aspect relates to a method for predicting severity of dysbiosis after treatment of a subject with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in a fecal sample or gut microbiota sample obtained from the subject. In embodiments, the fecal sample or gut microbiota sample is obtained from the subject prior to or during antibiotic treatment, preferably prior to antibiotic treatment.

Preferably, the invention provides a method for predicting severity of dysbiosis after treatment of a subject with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject. In embodiments, the fecal sample or gut microbiota sample is obtained from the subject prior to or during antibiotic treatment, preferably prior to antibiotic treatment.

In embodiments, the method comprises: (a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject; (b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably wherein the reference value representing a known severity of dysbiosis of a reference subject after treatment with an antibiotic; (c) finding a deviation or no deviation of the quantity of the bacteria as measured in (a) from said reference value; and (d) attributing the finding of deviation or no deviation to a particular prediction of severity of dysbiosis after treatment with an antibiotic. Preferably the reference value represents the quantity of said bacteria in a sample of a reference subject having developed a known severity of dysbiosis after treatment with an antibiotic. In embodiments, the method is for predicting severity of dysbiosis after treatment of a subject with a particular antibiotic or a particular class of antibiotics.

As shown in the experimental section, the present inventors have found that the present methods allow to predict the post-treatment microbial community type a patient is likely to develop after treatment with a particular class of antibiotics such as a combination of a penicillin and a betalactamase inhibitor; a beta-lactam antibiotic; or a fluoroquinolone, or after treatment with a particular antibiotic such as meropenem, a combination of amoxicillin and clavulanic acid, a combination of ampicillin and sulbactam, a combination of piperacillin and tazobactam, ciprofloxacin, levofloxacin, or ceftriaxone, and hence allow to select the particular class(es) of antibiotics or particular antibiotic(s) for treatment of a patient.

Hence, a further aspect of the invention provides method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in a fecal sample or gut microbiota sample obtained from the subject prior to (the envisaged) antibiotic treatment.

Preferably, the invention relates to a method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject prior to (the envisaged) antibiotic treatment.

The present methods advantageously allow to make correct treatment choices, such as early on after diagnosis of the infection, thereby improving patient quality of life. Incorrect treatment choices often lead to antibiotic-associated diarrhea, septicemia and even colitis-related death - a high cost to patients, healthcare system and insurances. The present methods hence also contribute to reduce costs in the healthcare sector.

In embodiments, the method comprises: (a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject prior to (the envisaged) antibiotic treatment; (b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably the reference value representing a reference subject having severe dysbiosis after treatment with the particular antibiotic or class of antibiotics; and (c) selecting the particular antibiotic or class of antibiotics for treatment of the subject if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is decreased or increased compared with the reference value. Preferably the reference value represents the amount of said bacteria in a sample of a reference subject as defined herein.

In particular embodiments, the method comprises: (a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject prior to (the envisaged) antibiotic treatment; (b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably the reference value representing a reference subject having severe dysbiosis after treatment with the particular antibiotic or class of antibiotics; and (c) selecting the particular antibiotic or class of antibiotics for treatment of the subject if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is increased compared with the reference value. Preferably the reference value represents the amount of said bacteria in a sample of a reference subject having developed severe dysbiosis after treatment with the particular antibiotic or class of antibiotics.

In embodiments, the method comprises detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject.

In embodiments, the method comprises detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject.

In embodiments, the method further comprises determining the level of expression of one or more antibiotic resistance genes in a fecal sample or gut microbiota sample obtained from the subject.

In embodiments, the detection of the bacteria is performed prior to or during antibiotic treatment; preferably the detection of the bacteria is performed prior to antibiotic treatment.

In embodiments, the reference value is a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria prior to antibiotic treatment of a reference subject or an average value of reference subjects.

In embodiments, the Faecalibacterium bacteria are Faecalibacterium prausnitzii; the Casaltella bacteria are Casaltella massiliensis; the Oscillibacter bacteria are selected from the group consisting of Oscillibacter massiliensis, Oscillibacter ruminantium, Oscillibacter valericigenes, Oscillibacter avistercoris, Oscillibacter excrementavium, Oscillibacter excrementigallinarum, and Oscillibacter pullicola; the Akkermansia bacteria are Akkermansia muciniphila; the Porphyromonas bacteria are Porphyromonas bennonis; and/or the Bifidobacterium bacteria are selected from the group consisting of Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, and Bifidobacterium catenulatum.

In embodiments, the step of detecting the bacteria may be carried out using nucleic acid sequencing such as RNA or DNA sequencing; quantitative polymerase chain reaction (qPCR); reverse transcription polymerase chain reaction (RT-PCR); polymerase chain reaction (PCR); digital PCR; rolling circle amplification (RCA); loop-mediated isothermal amplification (LAMP); a microarray; mass spectrometry; Western blot; immunohistochemistry; enzyme-linked immunosorbent assay (ELISA); or any combination of these methods. In embodiments, the step of comparing the quantity of the bacteria with a reference value may be performed using machine learning, linear discriminant analysis, linear regression, Spearman rank correlation, Euclidean distance; Manhattan distance; Average dot product; Pearson correlation; Pearson uncentered; Pearson squared; Cosine correlation; Covariance value; Kendall's Tau; or Mutual information.

In embodiments, the fecal sample may be a rectal swab or a stool sample.

In embodiments, the antibiotic or class of antibiotics is selected from the group consisting of betalactams, beta-lactam and beta-lactamase inhibitor combinations, penicillins, penicillin and betalactamase inhibitor combinations, penicillinase-resistant penicillins, penicillinase-resistant penicillin and beta-lactamase inhibitor combinations, cephalosporins, cephalosporin and betalactamase inhibitor combinations, carbapenems, carbapenem and beta-lactamase inhibitor combinations, monobactams, quinolones, fluoroquinolones, sulfonamides, aminoglycosides, tetracyclines, macrolides, glycopeptides, oxazolidinones, phenicols, lincosamides, Streptogramins, polymyxins, diaminopyrimidines, sulfones, para-aminobenzoic acid, bacitracin, isoniazid, rifamycins ethambutol, ethionamide, capreomycin, and clofazimine. In embodiments, the class of antibiotics may be a penicillin in combination with a beta-lactamase inhibitor, a beta-lactam antibiotic in combination with a beta-lactamase inhibitor, a beta-lactam antibiotic, or a fluoroquinolone antibiotic. In embodiments, the antibiotic may be ampicillin in combination with sulbactam; amoxicillin in combination with clavulanic acid; piperacillin in combination with tazobactam; ceftriaxone; meropenem; ciprofloxacin; or levofloxacin.

A further aspect of the invention relates to a kit of parts comprising a set of binding agents capable of detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in a fecal sample or gut microbiota sample obtained from a subject. Preferably, the invention provides a kit of parts comprising a set of binding agents capable of detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria in a fecal sample or gut microbiota sample obtained from a subject. Most preferably the binding agents capable of detecting bacteria in the kit consist of binding agents capable of selectively detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria in a fecal sample or gut microbiota sample obtained from a subject. In further embodiments, the kit comprises binding agents capable of selectively detection bacteria, said binding agents consisting of one or more probes capable of selectively detection one or more of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, or Bifidobacterium in a fecal or gut sample obtained from a subject. Preferably the kit comprises probes capable of selectively detecting two, three, four, five or all six of said bacteria.

Preferably, the invention provides a kit of parts comprising a set of binding agents capable of detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from a subject. Preferably, the invention provides a kit of parts comprising a set of binding agents capable of detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from a subject. Most preferably the binding agents capable of detecting bacteria in the kit consist of binding agents capable of detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from a subject.

In embodiments, the binding agent is polynucleotide probe capable of specifically binding to a nucleic acid of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria; preferably the binding agent is a set of oligonucleotides capable of specifically binding to a nucleic acid of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria. In embodiments, the binding agent is polynucleotide probe (or a set of probes) capable of specifically binding to a nucleic acid of the Faecalibacterium, Casaltella, and Oscillibacter bacteria and a polynucleotide probe (or a set of probes) capable of specifically binding to a nucleic acid of the one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria; preferably the binding agent is a set of oligonucleotides capable of specifically binding to a nucleic acid of the Faecalibacterium, Casaltella, and Oscillibacter bacteria and a nucleic acid of the one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria. The above and further aspects and preferred embodiments of the invention are described in the following sections and in the appended claims. The subject-matter of appended claims is hereby specifically incorporated in this specification.

DESCRIPTION OF THE DRAWINGS

Figure 1 represents a schematic overview of the patient and sample flow in the present study. The flow chart provides an overview of participating patients in each processing step, number of samples collected at each timepoint, and reasons for sample exclusion or non-collection. DI: rectal swab sample collected upon study enrolment. D6: rectal swab sample collected at the end of antibiotic treatment on day six ± 24 h or at hospital discharge. SAM: ampicillin/sulbactam. AMC: amoxicillin/clavulanic acid. TZP: piperacillin/tazobactam. CRO: ceftriaxone. MEM: meropenem. CIP: ciprofloxacin. LVX: levofloxacin.

Figure 2 represents a heatmap illustrating the unsupervised, de novo clustering of patient samples revealing two microbial community types (MCTs) post-treatment. Samples collected after broadspectrum antibiotic treatment (at D6, n = 310) were found to contain two distinct microbial communities (MCT1 and MCT2) when clustering samples de novo using the Dirichlet Multinomial Mixtures (DMM) algorithm. Compared to the composition of samples collected pre-treatment (at DI, n = 310), MCT2-like communities were characterized by ultra-low diversity and a distinct composition to that observed for MCTl-like communities. Differential abundant OTUs associated with each MCT were cross-sectionally identified by Linear Discriminant Analysis Effect Size (LEfSe, LDA > 3.0) at DI and D6 separately, and visualized using a heatmap. Only OTUs with a median relative abundance > 0.05% were considered. DI: rectal swab sample collected at study enrolment. D6: rectal swab sample collected at the end of antibiotic treatment on day six ± 24 h or at hospital discharge. LDA: Linear discriminant analysis score. OTU: operational taxonomical unit.

DETAILED DESCRIPTION OF THE INVENTION

As used herein, the singular forms "a", "an", and "the" include both singular and plural referents unless the context clearly dictates otherwise.

The terms "comprising", "comprises" and "comprised of" as used herein are synonymous with "including", "includes" or "containing", "contains", and are inclusive or open-ended and do not exclude additional, non-recited members, elements or method steps. The terms also encompass "consisting of" and "consisting essentially of".

The recitation of numerical ranges by endpoints includes all numbers and fractions subsumed within the respective ranges, as well as the recited endpoints.

The term "about" as used herein when referring to a measurable value such as a parameter, an amount, a temporal duration, and the like, is meant to encompass variations of and from the specified value, in particular variations of +/-10% or less, preferably +/-5% or less, more preferably +/-1% or less, and still more preferably +/-0.1% or less of and from the specified value, insofar such variations are appropriate to perform in the disclosed invention. It is to be understood that the value to which the modifier "about" refers is itself also specifically, and preferably, disclosed.

Whereas the term "one or more", such as one or more members of a group of members, is clear per se, by means of further exemplification, the term encompasses inter alia a reference to any one of said members, or to any two or more of said members, such as, e.g., any >3, >4, >5, >6 or >7 etc. of said members, and up to all said members.

All documents cited in the present specification are hereby incorporated by reference in their entirety.

Unless otherwise specified, all terms used in disclosing the invention, including technical and scientific terms, have the meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. By means of further guidance, term definitions may be included to better appreciate the teaching of the present invention.

As corroborated by the experimental section, which illustrates certain representative embodiments of the invention, the inventors realized that the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria prior to antibiotic treatment can be used as clinical markers for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic. Furthermore, said bacteria can be used as clinical markers for selecting a particular antibiotic or class of antibiotics for treatment of a subject.

Accordingly, a first aspect of the invention relates to a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, in a fecal sample or gut microbiota from the subject. Also provided in a related aspect is the use of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, as biomarkers useful for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic.

Further related aspects provide: a method for predicting severity of dysbiosis (that will develop) after treatment of a subject with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in a fecal sample or gut microbiota from the subject. the use of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria as biomarkers for predicting severity of dysbiosis that will develop after treatment of a subject with an antibiotic. a method for predicting severity of dysbiosis after treatment of a subject with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject. the use of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as biomarkers for predicting severity of dysbiosis after treatment of a subject with an antibiotic. a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment (i.e. if and when treated) with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, in a fecal sample or gut microbiota sample obtained from the subject prior to or during antibiotic treatment. a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment (i.e. if and when treated) with an antibiotic, the method comprising detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, in a fecal sample or gut microbiota sample obtained from the subject prior to or during antibiotic treatment. a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment (i.e. if and when) with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to Akkermansia, Porphyromonas, and Bifidobacterium bacteria, in a fecal sample or gut microbiota sample obtained from the subject prior to or during antibiotic treatment. a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment (i.e. if and when treated) with an antibiotic, the method comprising detecting Casaltella bacteria and Porphyromonas bacteria in addition to one or more of Faecalibacterium, Oscillibacter Akkermansia, or Bifidobacterium bacteria, in a fecal sample or gut microbiota sample obtained from the subject prior to or during antibiotic treatment. a method for predicting whether a subject is at risk of developing after treatment with an antibiotic (i.e. if and when treated) severe dysbiosis, the method comprising detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject prior to or during antibiotic treatment.

The inventors have shown that one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria can be used to predict whether a subject is at risk of developing severe dysbiosis. In particular, the inventors have found that the presence/quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria is different in subjects at risk of severe dysbiosis, e.g. as compared to the quantity of the same bacteria in subjects that are not at risk of severe dysbiosis, including in subjects who develop mild dysbiosis. In particular embodiments, the quantity of the one or more bacteria cited are lower (or decreased) in subjects at risk of severe dysbiosis.

In some embodiments of the methods or uses as taught herein, a difference in quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample compared to the reference value indicates that the subject will indicative for the risk of developing severe dysbiosis after treatment with an antibiotic.

In embodiments of the methods or uses as taught herein, a decreased or low quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample indicates that the subject will be at risk of developing severe dysbiosis after treatment with an antibiotic.

In certain embodiments, a normal quantity or increased or high quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample indicates that the subject will not be at risk of developing severe dysbiosis but will develop mild or moderate dysbiosis after treatment with an antibiotic. Such normal quantity of said bacteria or increased or decreased quantity of said bacteria may be assessed compared to a suitable reference value (i.e., a reference value of the quantity of each of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and, where appropriate each of the one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria) that represents one or more reference subjects (e.g., a population of reference subjects). In certain embodiments, the normal quantity of said bacteria may be assessed compared to a suitable reference value (i.e., a reference value of the quantity of each of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and where applicable each of the one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria) that represents the value in the sample of one or more reference subjects (e.g. a population of reference subjects) who have developed mild dysbiosis after treatment with an antibiotic, whereby a normal quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria may then refer to a quantity that is substantially the same as the reference value. The reference value corresponds to the quantity of said bacteria present in said reference subject(s) prior to said antibiotic treatment.

A subject determined or categorized as being at risk of developing severe dysbiosis after treatment (i.e. if and when being treated) with an antibiotic, for example a subject with an decreased (i.e. lower) quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample compared to a reference value representative of one or more subjects having developed mild dysbiosis after treatment with an antibiotic, may as a result of this analysis be treated with another antibiotic which is predicted to have a low impact on severity of dysbiosis or alternative treatment methods may be envisaged. The methods as described herein also allow, in subjects wherein the use of antibiotics is unavoidable, storing of stool to allow autologous fecal microbiota transplantation post-antibiotic treatment.

The phrases "predicting whether a subject is at risk of severe dysbiosis", "predicting the risk of severe dysbiosis", "determining risk of severe dysbiosis", or "determining whether a subject is at risk of severe dysbiosis" may be used interchangeably herein and all refer to determining the chances of developing the disease, prior to the start of severe dysbiosis. Also, the phrases "predicting mild dysbiosis" or "determining the likeliness of mild dysbiosis" or "determining whether a subject is likely to develop only mild dysbiosis" may be used interchangeably herein.

Similarly, the phrases "predicting severity of dysbiosis" or "determining the risk of a particular severity of dysbiosis" may be used interchangeably herein. The terms "predicting", "prediction" or "predictive" as used herein refers to an advance declaration, prognosis, indication or foretelling of a response or reaction to a therapy in a subject not (yet) having been treated with the therapy. For example, a prediction of mild dysbiosis to treatment with an antibiotic in a subject may indicate that the subject is eligible for treatment with an antibiotic, e.g., as the subject will have a clinical benefit from the treatment, e.g. without significant side effect on functioning of the gut. A prediction of risk of severe dysbiosis to treatment with an antibiotic in a subject may indicate that the subject is not eligible for treatment with an antibiotic, e.g., as the subject will not have a clinical benefit from the treatment or will suffer from significant side effects due to malfunctioning of the gut. Preferably, the subject will then be treated with a different antibiotic which is expected to result in mild dysbiosis. The term "prior to antibiotic treatment" or "prior to the envisaged antibiotic treatment" does not imply that the subject will be subjected to antibiotic treatment but merely indicates that the determination or selection is made before the decision to carry out said antibiotic treatment. In most embodiments, whether or not the subject is treated with said antibiotic will be determined by the outcome of said determination or selection.

The prediction prior to an envisaged treatment of a subject with an antibiotic may be performed on a sample (obtained) from a subject prior to or during antibiotic treatment of the subject. For instance, the envisaged treatment of the subject with the antibiotic may be the first treatment of an administration scheme. Alternatively, the envisaged treatment of the subject with the antibiotic may be one of the second or further treatments of an administration scheme or a continued treatment of a continuous administration regimen. For instance, the prediction prior to the envisaged second treatment of a subject with an antibiotic may be performed in a fecal sample or gut microbiota sample (obtained) from a subject after the first but before the second treatment of the subject, i.e. during treatment of the subject, with an antibiotic. For instance, the prediction prior to the envisaged second treatment of a subject with an antibiotic may be performed in a fecal sample or gut microbiota sample (obtained) from a subject during early antibiotic treatment, such as in a fecal sample or gut microbiota sample (obtained) from a subject within 48 hours or within 24 hours after intake of the first antibiotic dose.

The phrases "envisaged treatment of a subject with an antibiotic" or "the envisaged antibiotic treatment" as used herein refers to a potential treatment of the subject with an antibiotic. In other words, an envisaged treatment of a subject with an antibiotic will only become an actual treatment of the subject with the antibiotic when the outcome of the method as taught herein allows for the decision that the subject is to be treated with the antibiotic, e.g., without the risk of developing severe dysbiosis.

The phrase "prior to the envisaged treatment of a subject with an antibiotic" may encompass prior to the envisaged start of administration of the antibiotic, and prior to the envisaged continuation of the administration of the antibiotic.

The recitation "predicting whether the subject is at risk of developing severe dysbiosis" may encompass monitoring whether the subject is at risk of developing severe dysbiosis. The term "monitoring" generally refers to predicting whether the subject is at risk of developing severe dysbiosis over time. For instance, monitoring predicting whether the subject is at risk of developing severe dysbiosis may be performed by predicting whether the subject is at risk of developing severe dysbiosis at one or more successive time points.

The term "dysbiosis" as intended herein refers to dysbiosis of the gastrointestinal tract, in particular dysbiosis of the gut.

Generally, the term "dysbiosis" refers to any change to the components of resident gut commensal bacterial communities relative to the community found in healthy individuals. Dysbiosis can be categorized into three types that are not mutually exclusive (i) loss of beneficial microbial organisms, (ii) expansion of pathobionts or potentially harmful microorganisms, and (iii) loss of overall microbial diversity. For example, dysbiosis can be defined as a reduction in microbial diversity and a combination of the loss of beneficial bacteria such as Bacteroides strains and butyrate-producing bacteria such as Firmicutes and a rise in pathobionts (symbiotic bacteria that become pathogenic under certain conditions), including Proteobacteria, which encompasses gramnegative Escherichia coli.

The terms "dysbiosis", "dysbacteriosis", "gut microbiota dysbiosis" or "intestinal microbiota dysbiosis" as used interchangeably herein refer to a loss of beneficial microbial organisms and a loss of overall microbial diversity in the gut of a subject.

The phrase "severity of dysbiosis" refers to the degree of the loss of beneficial microbial organisms and the loss of overall microbial diversity in the gut of a subject. It will be understood that the severity of dysbiosis herein includes the options of severe dysbiosis, mild dysbiosis, or no dysbiosis. The terms "mild dysbiosis" or "moderate dysbiosis" refer to a low to moderate degree of loss of beneficial microbial organisms and loss of overall microbial diversity in the gut of a subject, e.g., without the occurrence of related adverse events such as AAD and C. difficile infection. For example, mild dysbiosis may occur for the duration of an antibiotic treatment, with the gut microbiota restoring after the treatment. The terms "severe dysbiosis" or "severe microbial perturbations" refer to a high degree of loss of beneficial microbial organisms and loss of overall microbial diversity in the gut of a subject, e.g., with the occurrence of related adverse events such as AAD and C. difficile infection. For example, severe dysbiosis may start by treatment with an antibiotic and cause long-lasting changes to the gut microbiota after the treatment resulting in adverse events.

The methods as disclosed herein may allow to make a prediction that a subject will be at risk of developing severe dysbiosis after treatment with an antibiotic. This may in certain embodiments include predicting that a subject will have a comparatively low probability (e.g., less than 50%, less than 40%, less than 30%, less than 20% or less than 10%) of developing severe dysbiosis after treatment with an antibiotic; or that a subject will have a comparatively high probability (e.g., at least 50%, at least 60%, at least 70%, at least 80% or at least 90%) of developing severe dysbiosis after treatment with an antibiotic.

In embodiments of the methods as taught herein, the methods may comprise determining, based on the presence in the sample of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium, that the subject is not at risk of developing severe dysbiosis. In embodiments of the methods as taught herein, the methods may comprise determining, based on the presence in the sample of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium, that the subject will develop low or mild dysbiosis. In embodiments of the methods as taught herein, the methods may comprise determining, based on the absence in the sample of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium, that the subject is at risk of developing severe dysbiosis.

Hence, an aspect of the invention provides a method for predicting whether the subject is at risk of developing severe dysbiosis after treatment with the antibiotic, the method comprising: detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject, in particular obtained from the subject prior to or during antibiotic treatment; and determining that the subject is not at risk (is at risk) of developing severe dysbiosis based on the presence (the absence) in the sample of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, such as based on the presence (the absence) in the sample of Faecalibacterium, Casaltella, and Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, or based on the presence (the absence) in the sample of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to Akkermansia, Porphyromonas, and Bifidobacterium bacteria, e.g., based on the presence (the absence) in the sample of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria.

A further aspect provides a method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, in a fecal sample or gut microbiota from the subject.

Also provided in a related aspect is the use of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, as biomarkers useful for selecting a particular antibiotic or class of antibiotics for treatment of a subject.

In certain embodiments of the methods or uses as taught herein, a decreased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample indicates that the subject will be at risk of developing severe dysbiosis after treatment with the particular antibiotic or class of antibiotics, and hence indicates not to select the particular antibiotic or class of antibiotics for treatment of the subject. In embodiments, a normal quantity or increased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample indicates that the subject will develop mild or moderate dysbiosis after treatment with the particular antibiotic or class of antibiotics, and hence indicates that the particular antibiotic or class of antibiotics can be selected for treatment of the subject. Such normal quantity of said bacteria or increased or decreased quantity of said bacteria may be assessed compared to a suitable reference value (i.e., a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria) that represents one or more reference subjects (e.g. a population of reference subjects) as taught herein.

Hence, an aspect of the invention provides a method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising: detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject, in particular prior to antibiotic treatment; and selecting (not selecting) the particular antibiotic or class of antibiotics for treatment of the subject based on the presence (the absence) in the sample of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, such as based on the presence (the absence) in the sample of Faecalibacterium, Casaltella, and Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, or based on the presence (the absence) in the sample of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to Akkermansia, Porphyromonas, and Bifidobacterium bacteria, e.g., based on the presence (the absence) in the sample of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria.

A subject determined or categorized as being at risk of developing severe dysbiosis after treatment with a particular antibiotic or class of antibiotics, for example a subject with an decreased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample compared to a reference value representative of a subject developing mild dysbiosis after treatment with an antibiotic, may therefore not be eligible for treatment with the particular antibiotic or class of antibiotics, and/or may receive a different antibiotic or class of antibiotics. The present methods may thus allow to stratify patients with a specific set of gut microbes for treatment with a particular antibiotic or class of antibiotics or to stratify patients having an infection for treatment with a particular antibiotic or a particular class of antibiotics (e.g., an antibiotic of a particular class of antibiotics) or to predict an outcome of treatment with a particular antibiotic or class of antibiotics. Based on the prediction, the treatment of the infection can be initiated, continued, or adapted. In certain embodiments, the methods or uses as taught herein are useful for predicting an outcome of treatment with a particular antibiotic or class of antibiotics in a subject having an infection. In embodiments, the outcome of treatment may be mild dysbiosis or severe dysbiosis.

The term "selecting" refers to choosing one or more items from a number or group of items. A selection can be made by excellence or arbitrarily.

The term "selecting an antibiotic" as used herein refers to choosing one or more antibiotic as being the best or most suitable from a number of antibiotics.

The term "selecting a class of antibiotics" refers to choosing one or more classes of antibiotics as being the best or most suitable from a number of classes of antibiotics. Once a class of antibiotics is selected, the method may comprise selecting an antibiotic as being the best or most suitable from the class of antibiotics or arbitrarily selecting an antibiotic from the class of antibiotics as all antibiotics from the class are known to be suitable.

In certain embodiments, the methods or uses as taught herein are useful for the stratification of subjects having an infection into groups for treatment with a particular antibiotic or class of antibiotics. Hence, a population of subjects having an infection may be stratified, i.e., divided or separated into subgroups or strata, based on the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, in samples from the subjects, or based on the severity of dysbiosis determined on the basis of said quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria. In certain embodiments, a subject may be allocated or classified to a given subgroup or stratum when the subject displays a quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria or a severity of dysbiosis corresponding to or encompassed by said subgroup or stratum. The subgroups or strata may each represent a treatment with a particular antibiotic or with an antibiotic of a particular class of antibiotics.

In certain embodiments, the methods or uses as taught herein are useful for guiding treatment of a subject with a particular antibiotic or class of antibiotics, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, in a fecal sample or gut microbiota from the subject. A related aspect provides the use of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria, optionally in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, as biomarkers useful for guiding treatment of a subject with a particular antibiotic or class of antibiotics.

In certain embodiments, the methods or uses as taught herein are useful for indicating treatment with a particular antibiotic or with a particular class of antibiotic as a suitable or unsuitable treatment for an infection in a subject.

In embodiments, a decreased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample, such as in particular a quantity of said bacteria that is lower than the reference value representative of mild dysbiosis after treatment with the particular antibiotic or class of antibiotics, indicates treatment with the particular antibiotic or class of antibiotics as an unsuitable treatment (as the subject is at risk of developing severe dysbiosis). In embodiments, a normal quantity or increased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample, such as in particular a quantity of said bacteria that is substantially the same as or higher than the reference value representative of mild dysbiosis after treatment with the particular antibiotic or class of antibiotics, indicates treatment with the particular antibiotic or class of antibiotics as a suitable treatment (as the subject is not at risk of developing severe dysbiosis).

In embodiments, the invention provides a method for predicting in a subject severity of dysbiosis after treatment of the subject with an antibiotic. In embodiments, the severity of dysbiosis after treatment of the subject with an antibiotic is predicted in the subject prior to treatment with the antibiotic. The terms "subject", "individual" or "patient" are used interchangeably throughout this specification, and typically and preferably denote humans, but may also encompass reference to non-human animals, preferably warm-blooded animals, even more preferably mammals, such as, e.g., non-human primates, rodents, canines, felines, equines, ovines, porcines, and the like. The term "non-human animals" includes all vertebrates, e.g., mammals, such as non-human primates, (particularly higher primates), sheep, dog, rodent (e.g., mouse or rat), guinea pig, goat, pig, cat, rabbits, cows, and non-mammals such as chickens, amphibians, reptiles etc. In certain embodiments, the subject is a non-human mammal. In certain preferred embodiments of the methods or uses as taught herein, the subject is a human subject. In other embodiments, the subject is an experimental animal or animal substitute as a disease model. The term does not denote a particular age or sex. Thus, adult and newborn subjects, as well as fetuses, whether male or female, are intended to be covered. Examples of subjects include humans, dogs, cats, cows, goats, and mice. The term subject is further intended to include transgenic species.

Suitable subjects may include without limitation subjects presenting to a physician for a screening for an infection, subjects presenting to a physician with symptoms and signs indicative of an infection, subjects diagnosed with an infection, subjects prior to antibiotic therapy, subjects undergoing antibiotic treatment, and subjects who have received antibiotic therapy.

The subject can be a subject that has an infection or has been diagnosed with an infection. The subject can be a subject that has been diagnosed with an infection and who receives, or will receive, an antibiotic for the treatment of said infection. Thanks to the method of the present invention, prediction can be made whether the subject who will receive the antibiotic treatment will be potentially at risk of developing severe dysbiosis or related adverse events such as antibiotic- associated diarrhea (AAD) as a consequence of said treatment. A prediction of a potential risk of developing severe dysbiosis can allow managing the subject with knowledge of this information. Such management can include, for example, further explorations of the most suitable antibiotic to prescribe to the subject to avoid the development of severe dysbiosis, the administration together with the antibiotic of a treatment designed to protect/preserve the gut microbiota during such treatment (this could consist in the administration of an enzyme to hydrolyse antibiotic residues in situ in the gut, or of an adsorbent to sequester antibiotic residues in the gut), administration of a microbiota complementation or replacement therapy (such as a fecal microbiota transplant, or one or several bacterial strains extracted from natural sources, or laboratory cultured and formulated) to complement or restore its microbiota following the antibiotic treatment, the isolation of the patient or admission of the subject into a clinical setting to monitor and handle the potential development of severe dysbiosis.

In embodiments, the subject is receiving or will receive at least one antibiotic. The subject can also be a patient receiving a prophylaxis with antibiotics to avoid a bacterial infection for example when the patient is immuno-compromised. Just like above, the detection of a high risk of severe dysbiosis could help healthcare providers adapt the care offered to the patient.

The subject can also be a patient with a history of severe dysbiosis or related adverse events such as AAD in the medical history. Such a patient is known to be at risk, and it could be interesting to detect patients at even higher risk of severe dysbiosis. Just like above, the detection of a high risk of severe dysbiosis could help healthcare providers adapt the care offered to the patient.

The subject can also be a patient that is screened to be enrolled in a clinical study to assess the efficacy of a drug or medical device to prevent severe dysbiosis. The prediction of a potential risk of developing severe dysbiosis will help decide if the patient is to be enrolled or not in the study given the objectives of clinical demonstration in the study.

Practice of the methods as taught herein requires a fecal sample from the subject or a gut microbiota sample from the subject, such as a sample which is taken prior to the (envisaged) treatment of said subject with antibiotics.

The terms "sample" or "biological sample" can be used interchangeably herein and encompass a fecal sample or gut microbiota sample obtained (isolated or removed) from a subject.

In embodiments of the methods as taught herein, the fecal sample may be a rectal swab or a stool sample.

The terms "gut microbiota" or "gut microbiota sample" may be used interchangeably herein and refer to a sample of the gut microbiota obtained from a subject. For example, the gut microbiota sample can be obtained through an ileostomy or obtained with a smart pill ingested by the patient and collecting gut microbiota content directly in the gut while travelling within the gastro-intestinal tract. The microbiota sample can be stored, for example to minimize exposure to air, in an adequate preparation media and/or frozen until further use in the methods as taught herein. For example, the fecal sample can be a stool sample or a rectal swab. These fecal samples can be obtained from a subject according to methods known in the art. Collection of a stool sample or rectal swab can be carried out in a clinical setting, or by the subject at home. The stool sample or rectal swab can be stored in an adequate preparation media or frozen until further use in the method of the present invention. The fecal sample or rectal swab can also be mailed by the patient to a laboratory that performs the methods as taught herein.

In embodiments, the sample may be a fecal sample or gut microbiota sample (obtained) from a subject being treated for or in need of treatment of an infection. In embodiments, the sample may be a fecal sample or gut microbiota sample (obtained) from a subject in need of treatment of an infection, wherein the sample is obtained from the subject prior to antibiotic treatment, e.g. prior to a treatment of the subject with any antibiotic or a particular antibiotic or an antibiotic of a particular class of antibiotics. In embodiments, the sample may be a fecal sample or gut microbiota sample (obtained) from a subject being treated with an antibiotic for an infection, wherein the sample is obtained from the subject during the treatment of the subject with an antibiotic. During the treatment of the subject with an antibiotic, the present methods allow to predict whether the subject is at risk of developing severe dysbiosis after continued treatment with the antibiotic, thereby allowing to monitor the risk of the subject over time and guide further treatment options.

In embodiments, the methods as taught herein comprise detecting one or more of Faecalibacterium, Clostridiales, or Oscillibacter bacteria in a fecal sample or gut microbiota from the subject. In embodiments, the methods as taught herein comprise detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in a fecal sample or gut microbiota from the subject. In embodiments, the methods as taught herein comprise detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

As described in the example section, Differential abundant OTUs were identified using Linear Discriminant Analysis Effect Size (LEfSe) identified bacterial taxa which were differentially expressed in reference subjects having severe dysbiosis and reference subjects having mild dysbiosis. Table 3 lists the identified bacterial taxa. Table 4 provides the identified bacterial species. The term "Faecalibacterium bacteria" as used herein refers to bacteria of the genus Faecalibacterium. The genus Faecalibacterium is annotated under the NCBI Taxonomy ID: 216851. The Faecalibacterium bacteria may be Faecalibacterium prausnitzii.

A non-limiting example of a Faecalibacterium prausnitzii strain has been described by Duncan et al. (International Journal of Systematic and Evolutionary Microbiology, 2002, 52, 2141-6). For example, this strain is deposited at the National Collection of Industrial, Food and Marine Bacteria (NCIMB) (Ferguson Building, Craibstone Estate, Bucksburn, Aberdeen, Scotland, UK) collection, accession number NCIMB 13872. This strain is available for example from the American Type Culture Collection (ATCC) (10801 University Boulevard, Manassas, Virginia, USA) public collection, accession number ATCC 27768. Faecalibacterium prausnitzii described by Duncan et al., 2002 is annotated under the NCBI Taxonomy ID: 853.

The term "Casaltella bacteria" as used herein refers to bacteria of the genus Casaltella. The genus Casaltella belongs to the class Clostridia. The genus Casaltella is annotated under the NCBI Taxonomy ID: 1715793. The Casaltella bacteria may be Casaltella massiliensis.

A non-limiting example of a Casaltella massiliensis strain has been described by La Scola et al. (Anaerobe, 2011, 17, 106-112). This strain is deposited for example at the Collection de Souches de I'Unite des Rickettsies (CSUR) (27 Blvd. Jean Moulin, Marseille, France), accession number CSUR P126. Casaltella massiliensis described by La Scola et al., 2011 is annotated under the NCBI Taxonomy ID: 938278.

The term "Oscillibacter bacteria" as used herein refers to bacteria of the genus Oscillibacter. The genus Oscillibacter is annotated under the NCBI Taxonomy ID: 459786. The Oscillibacter bacteria may be one or more of Oscillibacter massiliensis, Oscillibacter ruminantium, Oscillibacter valericigenes, Oscillibacter avistercoris, Oscillibacter excrementavium, Oscillibacter excrementigallinarum, and Oscillibacter pullicola. Preferably, the Oscillibacter bacteria is Oscillibacter massiliensis.

A non-limiting example of an Oscillibacter massiliensis strain has been described by Traore et al. (New Microbes New Infect., 2017, 19, 78-82). This strain is deposited for example at the Collection de Souches de I'Unite des Rickettsies (CSUR) (27 Blvd. Jean Moulin, Marseille, France), accession number CSUR P2778. Oscillibacter massiliensis described by Traore et al., 2017 is annotated under the NCBI Taxonomy ID: 1841866.

The term "Akkermansia bacteria" as used herein refers to bacteria of the genus Akkermansia. The genus Akkermansia is annotated under the NCBI Taxonomy ID: 239934. The Akkermansia bacteria may be Akkermansia muciniphila. A non-limiting example of an Akkermansia muciniphila strain has been described by Derrien et al. (Int. J. Syst. Evol. Microbiol., 2004, 54, 1469-1476). This strain is available for example from the ATCC public collection, accession number ATCC BAA-835. Akkermansia muciniphila described by Derrien et al., 2004 is annotated under the NCBI Taxonomy ID: 239935.

The term "Porphyromonas bacteria" as used herein refers to bacteria of the genus Porphyromonas. The genus Porphyromonas is annotated under the NCBI Taxonomy ID: 836. The Porphyromonas bacteria may be Porphyromonas bennonis.

A non-limiting example of a Porphyromonas bennonis strain has been described by Summanen et al. (Int. J. Syst. Evol. Microbiol., 2009, 59, 1727-1732). This strain is available for example from the ATCC public collection, accession number ATCC BAA-1629. Porphyromonas bennonis described by Summanen et al., 2009 is annotated under the NCBI Taxonomy ID: 501496.

The term "Bifidobacterium bacteria" as used herein refers to bacteria of the genus Bifidobacterium. The genus Bifidobacterium is annotated under the NCBI Taxonomy ID: 1678. The Bifidobacterium bacteria may be one or more of Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, and Bifidobacterium catenulatum.

A non-limiting example of a Bifidobacterium adolescentis strain has been described by Reuter (Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. I, 1963, 191, 486-507). This strain is available for example from the ATCC public collection, accession number ATCC:15703. Bifidobacterium adolescentis described by Reuter, 1963 is annotated under the NCBI Taxonomy ID: 1680.

In embodiments of the methods as taught herein, the Faecalibacterium bacteria may be Faecalibacterium prausnitzii; the Casaltella bacteria may be Casaltella massiliensis; the Oscillibacter bacteria may be selected from the group consisting of Oscillibacter massiliensis, Oscillibacter ruminantium, Oscillibacter valericigenes, Oscillibacter avistercoris, Oscillibacter excrementavium, Oscillibacter excrementigallinarum, and Oscillibacter pullicola; the Akkermansia bacteria may be Akkermansia muciniphila; the Porphyromonas bacteria may be Porphyromonas bennonis; and/or the Bifidobacterium bacteria may be selected from the group consisting of Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, and Bifidobacterium catenulatum.

In embodiments, the methods as taught herein comprise detecting one or more of Faecalibacterium prausnitzii, Casaltella massiliensis, Oscillibacter massiliensis, Oscillibacter ruminantium, Oscillibacter valericigenes, Oscillibacter avistercoris, Oscillibacter excrementavium, Oscillibacter excrementigallinarum, and Oscillibacter pullicola; optionally in addition to one or more of Akkermansia muciniphila, Porphyromonas bennonis, Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, and Bifidobacterium catenulatum.

In embodiments, the methods as taught herein may comprise measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in the fecal sample or gut microbiota from the subject. In embodiments, the methods as taught herein may comprise measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and (i.e., in addition to) one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota from the subject.

Bacteria or bacterial nucleic acid, peptide, polypeptide, or protein is "detected" or "measured" in a sample when the presence, absence and/or quantity (including the relative quantity or abundance) of said bacteria or said bacterial nucleic acid, peptide, polypeptide, or protein is determined or measured in the sample, preferably substantially to the exclusion of other bacteria or bacterial nucleic acids, peptides, polypeptides, or proteins.

In embodiments, the methods as taught herein comprise detecting the presence, absence and/or quantity of bacteria with a specific nucleic acid sequence belonging to one or more of Faecalibacterium, Casaltella, or Oscillibacter in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium in a fecal sample or gut microbiota from the subject. In embodiments, the methods as taught herein comprise detecting the relative quantity of said bacteria.

The terms "quantity", "amount" or "level" are synonymous and generally well-understood in the art. The terms as used herein may particularly refer to an absolute quantification of bacteria or bacterial nucleic acid, peptide, polypeptide, or protein in a sample, or to a relative quantification of bacteria or bacterial nucleic acid, peptide, polypeptide, or protein in a sample, i.e., relative to another value such as relative to a reference value as taught herein, or to a range of values indicating a baseline of the marker. These values or ranges may be obtained from a single patient or from a group of patients.

An absolute quantity of bacteria or bacterial nucleic acid, peptide, polypeptide, or protein in a sample may be advantageously expressed as weight or as molar amount, or more commonly as a concentration, e.g., weight per volume or mol per volume.

A relative quantity or relative abundance of bacteria or bacterial nucleic acid, peptide, polypeptide, or protein in a sample may be advantageously expressed as an increase or decrease or as a foldincrease or fold-decrease relative to another value, such as relative to a reference value as taught herein. Performing a relative comparison between first and second parameters (e.g., first and second quantities) may but need not require determining first the absolute values of said first and second parameters. For example, a measurement method may produce quantifiable readouts (such as, e.g., signal intensities) for said first and second parameters, wherein said readouts are a function of the value of said parameters, and wherein said readouts may be directly compared to produce a relative value for the first parameter vs. the second parameter, without the actual need to first convert the readouts to absolute values of the respective parameters.

Any existing, available or conventional separation, detection and/or quantification methods may be used to measure the presence or absence (e.g., readout being present vs. absent; or detectable amount vs. undetectable amount) and/or quantity (e.g., readout being an absolute or relative quantity) of bacteria or bacterial nucleic acid, peptide, polypeptide, or protein in a sample.

In embodiments of the methods as taught herein, the step of detecting the bacteria may be carried out using nucleic acid sequencing such as DNA or RNA sequencing; quantitative polymerase chain reaction (qPCR); reverse transcription polymerase chain reaction (RT-PCR); (highly multiplexed) polymerase chain reaction (PCR); digital PCR; rolling circle amplification (RCA); loop-mediated isothermal amplification (LAMP); a microarray; mass spectrometry; Western blot; immunohistochemistry; enzyme-linked immunosorbent assay (ELISA); or any combination of these methods. These techniques are performed as known in the art.

Detection of a bacterial genus or species or measure of the presence, absence and/or quantity of a bacterial genus or species can be carried out by any method known to those skilled in the art, for example and without limitation all the methods mentioned in Song et al. (Journal of Microbiology, 2018, 56(10), 693-705) or in Fraher et al. (Nature Reviews Gastroenterology & Hepatology, 2012, 9(6), 312-322). In some embodiments, the detection can be performed by culture, i.e., isolation of bacteria in selective media. In some embodiments, it can comprise detecting or measuring the level of a DNA, RNA, or protein unique to the bacterial genus or species of interest and can rely on techniques such as PCR, qPCR, DGGE, T-RFLP, FISH or DNA microarrays. In some embodiments, the detection or measure can be performed by sequencing of 16S rRNA gene of the bacteria, such as the V3-V4 or other regions, or it can be performed by shotgun sequencing or by targeted sequencing of specific genes. NextGen Sequencing (NGS) techniques can also be used as those described in Chiu & Miller, (Nature Reviews Genetics, 2019, 20, 341-355) or any technique that allows detection of bacterial genera, species, or a chromosomal gene specific to the bacterial or species including but not restricted to antibiotic resistance determinants. Isolation and analysis of nucleic acids (e.g., DNA, RNA) and proteins or other molecules from bacteria or produced by bacteria present in gut microbiota samples or fecal samples can be performed using established techniques that are known in the art and routinely used. For example, sample preparations in particular DNA extraction can be operated with the methods presented in Lim et al. (Systematic and Applied Microbiology, 2018, 41(2), 151-157) and in Costea et al. (Nat. Biotechnol., 2017). In particular, the DNA or RNA extraction can be performed with kits commercially available, compatible with the techniques contemplated for the next steps of analysis. Once DNA or mRNA is extracted from a sample, the amount of a bacterial DNA from a gene or a portion of gene whose sequence is unique to a bacterial species or strain, or the amount of RNA transcribed from a bacterial gene whose sequence or portions thereof are unique to a bacterial species or strain may be quantified. The preferred method for determining the DNA or RNA level is an amplification-based method, such as by polymerase chain reaction (PCR), including reverse transcription-polymerase chain reaction (RT-PCR) for RNA quantitative analysis, and detection by an appropriate method known in the art.

General molecular biology methods of nucleic acids extraction are known in the art.

The nucleic acids may also be obtained through in vitro amplification methods such as those described herein and known in the art. In some embodiments, the nucleic acids will not be amplified before they are quantified.

In embodiments, nucleic acid hybridization and/or amplification methods are used to detect and quantify nucleic acid sequences corresponding to specific bacterial groups that are to be detected or quantified in the methods as taught herein. In embodiments, an immunoassay or other assay to detect or quantify one or more specific proteins determinative of one or more of the bacteria can be used. For example, solid-phase ELISA immunoassays, Western blots, or immunohistochemistry are routinely used to specifically detect a protein. See Harlow and Lane Antibodies, A Laboratory Manual, Cold Spring Harbor Publications, NY (1988) for a description of suitable immunoassay formats and conditions.

DNA sequencing can be performed using known sequencing methodologies. Typically, a sample is sequenced using a large-scale sequencing method that provides the ability to obtain sequence information from many reads. Such sequencing platforms include for example those commercialized by Roche 454 Life Sciences (GS systems), Illumina (e.g., HiSeq, MiSeq), Life Technologies (e.g., SOLID systems), Thermo Fisher Scientific (Ion Torrent), Oxford Nanopore

Technologies (e.g., MinlON), and Pacific Biosciences (e.g., RSI I, Sequel). 1

Short-read sequencing technologies such as Roche 454 Life Sciences, Illumina, and Thermo Fisher Scientific technologies relies on the sequencing by synthesis principle. This involves either an emulsion PCR where DNA fragments are immobilized onto beads or an on-chip bridge amplification where DNA fragments hybridizes to a planar surface. Subsequent incorporation of bases is detected using fluorescence or ionic discharge. Methods that employ sequencing by hybridization may also be used. Such methods, e.g., used in the Life Technologies S0LiD4+ technology uses a pool of all possible oligonucleotides of a fixed length, labelled according to the sequence. Oligonucleotides are annealed and ligated; the preferential ligation by DNA ligase for matching sequences results in a signal informative of the nucleotide at that position.

The sequence can be determined using any other DNA sequencing method including, e.g., methods that use semiconductor technology to detect nucleotides that are incorporated into an extended primer by measuring changes in current that occur when a nucleotide is incorporated (see, e.g., U.S. Patent Application Publication Nos. 20090127589 and 20100035252). Other techniques include direct label- free exonuclease sequencing in which nucleotides cleaved from the nucleic acid are detected by passing through a nanopore (Oxford Nanopore) (Clark et al., Nature Nanotechnology 4: 265-270, 2009); and Single Molecule Real Time (SMRT™) DNA sequencing technology (Pacific Biosciences),

Deep sequencing can also be used to quantify the number of copies of a particular sequence in a sample and then also be used to determine the relative quantity of different sequences in a sample. Deep sequencing refers to sequencing of a nucleic acid sequence, for example such that the original number of copies of a sequence in a sample can be determined or estimated.

In some embodiments, specific sequences in the sample can be targeted for amplification and/or sequencing. For example, specific primers can be used to detect and sequence bacterial sequences of interest and corresponding to the bacterial species to detect in the method of the invention. In addition, or alternatively, whole genome sequencing methods that sequence random DNA fragments in a sample can be used.

Once raw sequencing data is generated, conventional methods for primary data analysis known to those skilled in the art may be conducted to trim sequences based on quality, remove ambiguous sequences and chimera, and cluster and classify reads. Resulting sequence reads can be further classified by the use of single-nucleotide resolution methods by comparing the resulting sequence reads to known sequences in a genomic database. Illustrative algorithms that are suitable for determining percent sequence identity and sequence similarity and thus aligning and identifying sequence reads are for example the BLAST and BLAST 2.0 algorithms. Accordingly, for the sequence reads generated, a subset of these reads can be aligned to one or more bacterial genomes of the bacterial species as taught herein. For example, one can align a read with a database of bacterial sequences and the read can be designated as from a particular bacteria if that read has the best alignment to a DNA sequence from that bacteria in the database.

In some embodiments, the genes of interest to identify the bacterial species as taught herein may be placed on DNA microarrays or DNA chips to permit a fast detection and measure of the bacterial species of interest. Microarray technology is a high throughput platform used to study numerous samples and to detect thousands of nucleic acid sequences simultaneously making it fast and user friendly. Phylogenetic DNA microarrays consist of several thousand probes, usually designed from rRNA gene sequence database targeting either specific organisms (e.g., pathogenic bacteria) or the whole microbiota at various taxonomic levels. Several microarrays addressing the gut microbiota have been developed over the last decade, showing differences in their design and the aims of study, for instance as described in WO2021/123387. In 2013, the Human Intestinal Tract Chip (HITChip) was designed to target 1,140 species using 4,809 overlapping probes. See Tottey et al. (PLoS ONE, 2013, 8(5): e62544). Microarray technologies may be used to develop devices capable of measuring the relative quantity of the bacterial species of the invention in a very quick fashion.

In a further embodiment, when detection or measure is performed by sequencing of 16S rRNA gene, taxonomic identification can be carried out using the SILVA high quality ribosomal RNA databases (https://www.arb-silva.de/), Greengenes or any other suitable ribosomal RNA database. In embodiments of the methods as taught herein, the methods may comprise detecting at least one, such as at least two or all three, of Faecalibacterium, Casaltella, or Oscillibacter bacteria and at least one, such as at least two or all three, of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota from the subject. For instance, the methods may comprise detecting at least two of Faecalibacterium, Casaltella, or Oscillibacter bacteria and at least one of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota from the subject. For instance, the methods may comprise detecting at least two of Faecalibacterium, Casaltella, or Oscillibacter bacteria and at least two of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota from the subject. In embodiments of the methods as taught herein, the methods may comprise detecting at least four, such as at least five or all six, of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria (or bacterial species) in a fecal sample or gut microbiota from the subject. In embodiments of the methods as taught herein, the methods may comprise detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria in addition to one or more of, such as at least one, at least two, or all three, of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Faecalibacterium, Casaltella, Oscillibacter, and Akkermansia bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Faecalibacterium, Casaltella, Oscillibacter, and Porphyromonas bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Faecalibacterium, Casaltella, Oscillibacter, and Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, and Porphyromonas bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, and Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Faecalibacterium, Casaltella, Oscillibacter, Porphyromonas, and Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject. As show in the examples, such combination of bacteria allows predicting whether a subject is at risk of severe dysbiosis prior to treatment of the subject with an antibiotic and/or selecting a particular antibiotic or class of antibiotics for treatment of a subject with sufficient sensitivity and specificity prior to treatment of the subject.

In embodiments of the methods as taught herein, the methods may comprise detecting Akkermansia, Porphyromonas, and Bifidobacterium bacteria and one or more, such as at least one, such as at least two or all three, of Faecalibacterium, Casaltella, or Oscillibacter bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Akkermansia, Porphyromonas, Bifidobacterium, and Faecalibacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Akkermansia, Porphyromonas, Bifidobacterium, and Casaltella bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Akkermansia, Porphyromonas, Bifidobacterium, and Oscillibacter bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Akkermansia, Porphyromonas, Bifidobacterium, Faecalibacterium, and Casaltella bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Akkermansia, Porphyromonas, Bifidobacterium, Faecalibacterium, and Oscillibacter bacteria in the fecal sample or gut microbiota sample obtained from the subject. In embodiments, the methods may comprise detecting Akkermansia, Porphyromonas, Bifidobacterium, Casaltella, and Oscillibacter bacteria in the fecal sample or gut microbiota sample obtained from the subject. As show in the examples, such combination of bacteria allows predicting whether a subject is at risk of severe dysbiosis prior to treatment of the subject with an antibiotic and/or selecting a particular antibiotic or class of antibiotics for treatment of a subject with sufficient sensitivity and specificity prior to treatment of the subject.

In embodiments of the methods as taught herein, the methods may comprise detecting Casaltella bacteria and Porphyromonas bacteria in addition to one or more of, such as at least one, at least two, at least three, or all four of, Faecalibacterium, Oscillibacter Akkermansia, or Bifidobacterium bacteria.

In embodiments of the methods as taught herein, the methods may comprise detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject. As show in the examples, such methods allow predicting whether a subject is at risk of severe dysbiosis prior to treatment of the subject with an antibiotic and/or selecting a particular antibiotic or class of antibiotics for treatment of a subject with sufficient sensitivity and specificity prior to treatment of the subject.

Hence, an aspect provides a method for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic, the method comprising detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

A further aspect provides a method for predicting severity of dysbiosis after treatment of a subject with an antibiotic, the method comprising detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

A further aspect provides a method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

Related aspects provide: the use of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria as biomarkers useful for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic. the use of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria as biomarkers useful for predicting severity of dysbiosis after treatment of a subject with an antibiotic. the use of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria as biomarkers useful for selecting a particular antibiotic or class of antibiotics for treatment of a subject.

For instance, the methods may comprise measuring the following combinations of bacterial species: Faecalibacterium prausnitzii; Casaltella massiliensis; Oscillibacter massiliensis; Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum.

Faecalibacterium prausnitzii; Casaltella massiliensis; Oscillibacter ruminantium; Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum.

Faecalibacterium prausnitzii; Casaltella massiliensis; Oscillibacter valericigenes; Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum.

Faecalibacterium prausnitzii; Casaltella massiliensis; Oscillibacter avistercoris; Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum.

Faecalibacterium prausnitzii; Casaltella massiliensis; Oscillibacter excrementavium; Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum.

Faecalibacterium prausnitzii; Casaltella massiliensis; Oscillibacter excrementigallinarum; Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum.

Faecalibacterium prausnitzii; Casaltella massiliensis; Oscillibacter pullicola; Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum.

In embodiments of the methods as taught herein, the detection of the bacteria may be performed prior to or during antibiotic treatment; preferably the detection of the bacteria is performed prior to antibiotic treatment. Such detection prior antibiotic treatment advantageously allows to determine whether the subject is at risk of developing severe dysbiosis prior to starting antibiotic therapy, and hence improves patient quality of life. Furthermore, prior to treatment with a particular antibiotic or class of antibiotics, the present methods advantageously allow to make correct treatment choices, thereby minimizing the impact of the antibiotic therapy on both the gut microbiome and the resistance gene reservoir.

In embodiments of the methods as taught herein, the detection of the bacteria may be performed in a fecal sample or gut microbiota sample (obtained) from the subject prior to or during antibiotic treatment. In embodiments of the methods as taught herein, the detection of the bacteria may be performed in a fecal sample or gut microbiota sample (obtained) from the subject prior to potential antibiotic treatment. In embodiments of the methods as taught herein, the detection of the bacteria may be performed in a fecal sample or gut microbiota sample (obtained) from the subject prior to deciding on the antibiotic treatment. In embodiments of the methods as taught herein, the detection of the bacteria may be performed in a fecal sample or gut microbiota sample (obtained) from the subject prior to deciding whether or not to treat the subject with the antibiotic.

In embodiments, the methods as taught herein may comprise comparing the quantity of the bacteria as measured in the sample from the subject with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria.

In embodiments, the quantity of each bacteria can be compared to a threshold value corresponding to a predetermined value under which or over which the measured quantity will be considered predictive or not predictive of a risk of developing severe dysbiosis. For example, a threshold value may correspond to a predetermined value representing subjects having severe dysbiosis, and a measure higher than such predetermined threshold value is indicative of developing mild dysbiosis and hence no risk of developing severe dysbiosis. On the contrary, a threshold value may correspond to a predetermined value representing subjects having mild dysbiosis, and a measure lower than a predetermined threshold value is indicative of a risk of developing severe dysbiosis.

In embodiments, the measures obtained for more than one bacteria as taught herein, such as two bacteria, can be used to determine a ratio. For example, the measure of the relative abundance of two bacterial species can be used to determine a ratio of relative abundances. In a particular embodiment, the ratio is calculated of the relative abundance of a bacterial species associated to subjects at risk of developing severe dysbiosis to the relative abundance of a bacterial species associated to subjects developing mild dysbiosis. In this embodiment, the lower the ratio is, the higher the risk is for the subject to develop severe dysbiosis. Further for example, the ratio may be calculated of the relative abundance of a bacterial species associated to subjects at risk of developing severe dysbiosis to the relative abundance of a bacterial species remaining similar or stable in subjects developing mild dysbiosis and severe dysbiosis. In an embodiment, the ratio calculated from the measures carried out from the gut microbiota sample or the fecal sample of the subject can be compared to a predetermined control ratio. Such control ratio can be set so that a calculated ratio lower than this control ratio is indicative of a risk of developing severe dysbiosis, while a calculated ratio higher than this control ratio is indicative of developing mild dysbiosis and hence no risk of developing severe dysbiosis. Such control ratio can also be set so that a calculated ratio higher than this control ratio is indicative of a risk of developing severe dysbiosis, while a calculated ratio lower than this control ratio is indicative of developing mild dysbiosis and hence no risk of developing severe dysbiosis.

In embodiments, the quantity (including the relative abundance) of each of the bacteria as taught herein measured in the sample may be compared with a reference value of the quantity (including the relative abundance) of each of the bacteria as taught herein in a reference subject.

The term "reference subject" as used herein encompasses one or more reference subjects such as a population of reference subjects.

In embodiments, the reference value may represent a reference subject having developed severe dysbiosis after treatment with an antibiotic ("any" antibiotic) or after treatment with a particular antibiotic or particular class of antibiotics. In embodiments, the reference value may represent the quantity of bacteria in a sample of a reference subject having developed severe dysbiosis after treatment with an antibiotic ("any" antibiotic) or after treatment with a particular antibiotic or particular class of antibiotics. The reference value may represent a reference subject having developed severe dysbiosis after treatment with an antibiotic or may represent a reference subject having developed mild dysbiosis after treatment with an antibiotic. Hence, in embodiments, the method comprises: (a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota from the subject; (b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, wherein the reference value represents a reference subject having severe dysbiosis after treatment with an antibiotic; (c) predicting that the subject is at risk (is not at risk) of developing severe dysbiosis after treatment with an antibiotic if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and optionally one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is the same or decreased (is increased) compared with the reference value.

In embodiments, the reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria (referred to herein as "the bacteria as taught herein") may correspond to the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the sample of a reference subject who has developed severe dysbiosis after antibiotic treatment or after treatment with a particular antibiotic or particular class of antibiotics. In embodiments, the reference value of the quantity the bacteria as taught herein may correspond to the quantity of the bacteria as taught herein in a reference subject who has mild dysbiosis after antibiotic treatment or after treatment with a particular antibiotic or particular class of antibiotics. In embodiments, the reference value of the quantity the bacteria as taught herein may correspond to the quantity of the bacteria as taught herein in a reference subject who has no dysbiosis after antibiotic treatment or after treatment with a particular antibiotic or particular class of antibiotics.

In embodiments, the methods as taught herein may rely on comparing the quantity of the bacteria as taught herein measured in a sample from the subject with reference values, wherein said reference values represent a known severity of dysbiosis after treatment with an antibiotic or after treatment with a particular antibiotic or particular class of antibiotics.

For example, a reference value of the quantity of the bacteria as taught herein may represent a known severity of dysbiosis after treatment with an antibiotic or after treatment with a particular antibiotic or particular class of antibiotics. For instance, a reference value of the quantity of the bacteria as taught herein may represent severe dysbiosis, mild dysbiosis, or no dysbiosis after treatment with an antibiotic or after treatment with a particular antibiotic or particular class of antibiotics.

Hence, in embodiments, the method may comprise:

(a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and of one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota from the subject;

(b) comparing the quantity of bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, the reference value representing a known severity of dysbiosis of a reference subject after treatment with an antibiotic;

(c) finding a deviation or no deviation of the quantity of the bacteria as measured in (a) from said reference value; and

(d) attributing the finding of deviation or no deviation to a particular prediction of severity of dysbiosis after treatment of the subject with an antibiotic.

In embodiments, said reference value represents a reference subject having severe dysbiosis after treatment with an antibiotic, and wherein: the same or a decreased quantity of the bacteria as measured in (a) compared with the reference value indicates that the subject will be at risk of developing severe dysbiosis after treatment with an antibiotic, or an increased quantity of the bacteria as measured in (a) compared with the reference value indicates that the subject will not be at risk of developing severe dysbiosis after treatment with an antibiotic.

In embodiments, said reference value represents a reference subject having mild dysbiosis after treatment with an antibiotic, and wherein: the same or an increased quantity of the bacteria as measured in (a) compared with the reference value indicates that the subject will not be at risk of developing severe dysbiosis after treatment with an antibiotic, or a decreased quantity of the bacteria as measured in (a) compared with the reference value indicates that the subject will be at risk of developing severe dysbiosis after treatment with an antibiotic.

Further, a reference value of the quantity of the bacteria as taught herein may represent a reference subject with a known treatment choice with a particular antibiotic or particular class of antibiotics. For instance, a reference value of the quantity of the bacteria as taught herein may represent a reference subject being eligible to treatment with a particular antibiotic or class of antibiotics, or a reference subject being not eligible to (e.g., excluded from) treatment with a particular antibiotic or class of antibiotics.

Hence, in embodiments, the method may comprise:

(a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota from the subject;

(b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, the reference value representing a known treatment choice of a reference subject after treatment with the particular antibiotic or class of antibiotics;

(d) attributing the finding of deviation or no deviation to a particular treatment choice of the subject with the particular antibiotic or class of antibiotic.

In embodiments, said reference value represents a reference subject that is eligible to treatment with a particular antibiotic or particular class of antibiotics, and wherein: the same or an increased quantity of the bacteria as measured in (a) compared with the reference value indicates treatment of the subject with the particular antibiotic or class of antibiotics, or a decreased quantity of the bacteria as measured in (a) compared with the reference value indicates no eligibility to (e.g., exclusion of) treatment of the subject with the particular antibiotic or class of antibiotics.

In embodiments, said reference value represents a reference subject that is not eligible to treatment with the particular antibiotic or particular class of antibiotics, and wherein: the same or a decreased quantity of the bacteria as measured in (a) compared with the reference value indicates no eligibility (e.g., exclusion of) treatment of the subject with the particular antibiotic or class of antibiotics, or an increased quantity of the bacteria as measured in (a) compared with the reference value indicates treatment of the subject with the particular antibiotic or class of antibiotics. In embodiments, the reference values of the quantity of each of the bacteria as taught herein in a reference subject may be provided as a reference profile.

In embodiments, the quantity of each of the bacteria as taught herein measured in the sample may be compared with a reference profile comprising the reference values of the quantity of each of the bacteria as taught herein in a reference subject.

In embodiments, the reference profiles may be obtained from reference samples. In embodiments, the reference profiles may be obtained from reference samples which are obtained from reference subjects. A reference subject or group of reference subjects may have or may be known to have a particular microbial community type. A reference subject or group of reference subjects may be known to have (developed) severe dysbiosis after antibiotic treatment or to have mild dysbiosis after antibiotic treatment.

In embodiments, the comparison may be performed using machine learning, linear discriminant analysis, linear regression, Spearman rank correlation, Euclidean distance; Manhattan distance; Average dot product; Pearson correlation; Pearson uncentered; Pearson squared; Cosine correlation; Covariance value; Kendall's Tau; or Mutual information. In embodiments, the step of comparing the quantity of the bacteria with a reference value may be performed using machine learning, linear discriminant analysis, linear regression, Spearman rank correlation, Euclidean distance; Manhattan distance; Average dot product; Pearson correlation; Pearson uncentered; Pearson squared; Cosine correlation; Covariance value; Kendall's Tau; or Mutual information.

The comparison step of the methods as taught herein may generally include any means to determine the presence or absence of at least one difference or deviation and optionally of the size of such difference or deviation between values being compared. A comparison may include a visual inspection, an arithmetical or statistical comparison of measurements. Such statistical comparisons include, but are not limited to, applying a rule.

Reference values may be established according to known procedures. For example, a reference value may be established in a reference subject or individual or a population of individuals characterized by a particular severity of dysbiosis. Such population may comprise without limitation 2 or more, 10 or more, 100 or more, or even several hundred or more individuals.

A "deviation" of a first value from a second value may generally encompass any direction (e.g., increase: first value > second value; or decrease: first value < second value) and any extent of alteration.

For example, a deviation may encompass a decrease in a first value by, without limitation, at least about 10% (about 0.9-fold or less), or by at least about 20% (about 0.8-fold or less), or by at least about 30% (about 0.7-fold or less), or by at least about 40% (about 0.6-fold or less), or by at least about 50% (about 0.5-fold or less), or by at least about 60% (about 0.4-fold or less), or by at least about 70% (about 0.3-fold or less), or by at least about 80% (about 0.2-fold or less), or by at least about 90% (about 0.1-fold or less), relative to a second value with which a comparison is being made.

For example, a deviation may encompass an increase of a first value by, without limitation, at least about 10% (about 1.1-fold or more), or by at least about 20% (about 1.2-fold or more), or by at least about 30% (about 1.3-fold or more), or by at least about 40% (about 1.4-fold or more), or by at least about 50% (about 1.5-fold or more), or by at least about 60% (about 1.6-fold or more), or by at least about 70% (about 1.7-fold or more), or by at least about 80% (about 1.8-fold or more), or by at least about 90% (about 1.9-fold or more), or by at least about 100% (about 2-fold or more), or by at least about 150% (about 2.5-fold or more), or by at least about 200% (about 3-fold or more), or by at least about 500% (about 6-fold or more), or by at least about 700% (about 8-fold or more), or like, relative to a second value with which a comparison is being made.

Preferably, a deviation may refer to a statistically significant observed alteration. For example, a deviation may refer to an observed alteration, which falls outside of error margins of reference values in a given population (as expressed, for example, by standard deviation or standard error, or by a predetermined multiple thereof, e.g., ±lxSD or ±2xSD or ±3xSD, or ±lxSE or ±2xSE or ±3xSE). Deviation may also refer to a value falling outside of a reference range defined by values in a given population (for example, outside of a range which comprises >40%, > 50%, >60%, >70%, >75% or >80% or >85% or >90% or >95% or even >100% of values in said population).

In a further embodiment, a deviation may be concluded if an observed alteration is beyond a given threshold or cut-off. Such threshold or cut-off may be selected as generally known in the art to provide for a chosen sensitivity and/or specificity of the prediction methods, e.g., sensitivity and/or specificity of at least 50%, or at least 60%, or at least 70%, or at least 80%, or at least 85%, or at least 90%, or at least 95%.

For example, receiver-operating characteristic (ROC) curve analysis can be used to select an optimal cut-off value, e.g. of the quantity of the bacteria as taught herein, for clinical use of the methods as taught herein, based on acceptable sensitivity and specificity, or related performance measures which are well-known per se, such as positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR-), Youden index, or similar. By means of an example, a cut-off value may be selected such as to provide for AUC value higher than 50%, or higher than 55%, or higher than 60%, or higher than 65%, or higher than 70%, or higher than 75%, or higher than 80%, or higher than 85%, or higher than 90%, or higher than 95%.

In embodiments of the methods as taught herein, the reference value may be a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria during or prior to antibiotic treatment of the reference subject. Preferably, the reference value is a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria prior to antibiotic treatment of the reference subject. In embodiments of the methods as taught herein, the reference value may be a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria during or prior to antibiotic treatment of the reference subject. Preferably, the reference value is a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria prior to antibiotic treatment of the reference subject.

Mentioned herein, the present methods allow to predict whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic (e.g., "any" antibiotic), a particular class of antibiotics or even a particular antibiotic.

In embodiments, the methods as taught herein are for predicting whether a subject is at risk of developing severe dysbiosis after treatment with a particular class of antibiotics or a particular antibiotic. In embodiments, the methods as taught herein are for predicting severity of dysbiosis after or upon treatment of a subject with a particular class of antibiotics or a particular antibiotic. In embodiments, the methods as taught herein are for selecting a particular antibiotic or class of antibiotics for treatment of a subject. In embodiments, the methods as taught herein are for stratifying a subject for treatment with a particular antibiotic or class of antibiotics. In embodiments, the methods may comprise: (i) selecting the particular antibiotic or class of antibiotics for treatment of the subject if the subject is predicted to develop mild dysbiosis with a particular antibiotic or class of antibiotics, or (ii) not selecting the particular antibiotic or class of antibiotics for treatment of the subject if the subject is predicted to be at risk of developing severe dysbiosis with a particular antibiotic or class of antibiotics.

In embodiments, the method comprises: (a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota in a sample obtained from the subject prior to (the envisaged) antibiotic treatment; (b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably the reference value representing a reference subject having (developed) severe dysbiosis after treatment with the particular antibiotic or class of antibiotics; (c) selecting (not selecting) the particular antibiotic or class of antibiotics for treatment of the subject if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is increased (the same or decreased) compared with the reference value.

In embodiments, the methods as taught herein may comprise detecting one or more of Ruminococcaceae, Bacteroides, Clostridium XlVa, Prevotella, Bilophila, Campylobacter, Ruminococcus, Akkermansia, Porphyromonas, Bifidobacterium, Ruminococcus2, Mobiluncus, Alistipes, Blautia, Anaerococcus, Peptoniphilus, Finegoldia, Enterococcus, Phenylobacterium, Erysipelotrichaceae, Escherichia, Shigella, Sphingomonas, and Parabacteroides bacteria in a fecal sample or gut microbiota from the subject instead of or in addition to detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in a fecal sample or gut microbiota from the subject.

In embodiments, the methods as taught herein may comprise one or more prior steps of isolating a fecal sample or gut microbiota of the subject, providing a nucleic acid extract from a fecal sample or gut microbiota of the subject, and amplifying nucleic acid regions using DNA primers specific for the bacteria as taught herein. In embodiments, the methods as taught herein may comprise amplifying regions of the 16S rRNA genes such as the V3-V4 regions for instance by PCR and sequencing regions of the 16S rRNA genes such as the V3-V4 regions.

In embodiments of the methods as taught herein, the methods may further comprise determining the level of expression of one or more antibiotic resistance genes in a fecal sample or gut microbiota sample obtained from the subject. Such step may further contribute to the decision to select a particular antibiotic or class of antibiotics for treatment of a subject.

The antibiotic as described herein may refer to any antibiotic or to a particular antibiotic or to an antibiotic of a particular class of antibiotics.

In embodiments of the methods as taught herein, the antibiotic or class of antibiotics may be selected from the group consisting of beta-lactams, beta-lactam and beta-lactamase inhibitor combinations, penicillins, penicillin and beta-lactamase inhibitor combinations, penicillinaseresistant penicillins, penicillinase-resistant penicillin and beta-lactamase inhibitor combinations, cephalosporins, cephalosporin and beta-lactamase inhibitor combinations, carbapenems, carbapenem and beta-lactamase inhibitor combinations, monobactams, quinolones, fluoroquinolones, sulfonamides, aminoglycosides, tetracyclines, macrolides, glycopeptides, oxazolidinones, phenicols, lincosamides, Streptogramins, polymyxins, diaminopyrimidines, sulfones, para-aminobenzoic acid, bacitracin, isoniazid, rifamycins ethambutol, ethionamide, capreomycin, and clofazimine.

In embodiments of the methods as taught herein, the class of antibiotics may be a penicillin, a betalactam antibiotic, a beta-lactamase inhibitor, a fluoroquinolone antibiotic, or a combination thereof. In embodiments of the methods as taught herein, the class of antibiotics may be a penicillin in combination with a beta-lactamase inhibitor, a beta-lactam antibiotic in combination with a betalactamase inhibitor, a beta-lactam antibiotic, or a fluoroquinolone antibiotic. As show in the example section, the present methods allow predicting whether a subject is at risk of severe dysbiosis prior to treatment of the subject with a particular class of antibiotics, and thus selecting the particular class of antibiotics for treatment of a subject with sufficient sensitivity and specificity. In embodiments of the methods or uses as taught herein, the antibiotic may be ampicillin, sulbactam, amoxicillin, clavulanic acid, piperacillin, tazobactam, ceftriaxone, meropenem, ciprofloxacin, levofloxacin, or a combination thereof. In embodiments of the methods or uses as taught herein, the antibiotic may be ampicillin in combination with sulbactam; amoxicillin in combination with clavulanic acid; piperacillin in combination with tazobactam; ceftriaxone; meropenem; ciprofloxacin; or levofloxacin. As show in the example section, the present methods allow predicting whether a subject is at risk of severe dysbiosis prior to treatment of the subject with a particular antibiotic, and thereby allowing to select the particular antibiotic for treatment of a subject with sufficient sensitivity and specificity.

The present invention also provides compositions and methods for treating or preventing severe dysbiosis.

As mentioned above, the inventors have shown that certain bacteria are more abundant in subjects that will not develop severe dysbiosis. Accordingly, an aspect provides a composition comprising one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria. In an embodiment, the bacteria are one or more of Faecalibacterium prausnitzii; Casaltella massiliensis; and Oscillibacter massiliensis, Oscillibacter ruminantium, Oscillibacter valericigenes, Oscillibacter avistercoris, Oscillibacter excrementavium, Oscillibacter excrementigallinarum, or Oscillibacter pullicola; and one or more of is Akkermansia muciniphila; Porphyromonas bennonis; and Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, or Bifidobacterium catenulatum. Such compositions can be used in a method for the treatment or prevention of severe dysbiosis.

Hence an aspect provides one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria for use in a method of treating or preventing severe dysbiosis in a subject.

Related aspects provide: a method of treating or preventing severe dysbiosis in a subject in need of such a treatment, comprising administering a therapeutically effective amount of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria to the subject. the use of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria for the manufacture of a medicament for the treatment or prevention of severed dysbiosis in a subject.

The methods as taught herein further allow to treat a subject which has been determined not to be at risk of developing severe dysbiosis with an antibiotic or with a particular antibiotic or class of antibiotics.

Hence, a further aspect relates to an antibiotic for use in a method of treating an infection in a subject, wherein the subject has been selected as having an increased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject. In embodiments, the antibiotic may be ampicillin, sulbactam, amoxicillin, clavulanic acid, piperacillin, tazobactam, ceftriaxone, meropenem, ciprofloxacin, levofloxacin, or a combination thereof, such as ampicillin in combination with sulbactam; amoxicillin in combination with clavulanic acid; piperacillin in combination with tazobactam; ceftriaxone; meropenem; ciprofloxacin; or levofloxacin.

A further aspect relates to an antibiotic for use in a method of treating an infection in a subject, wherein the subject has been selected as not being at risk of developing severe dysbiosis by the methods as taught herein.

Further aspects provide: a method of treating an infection in a subject in need of such a treatment, comprising administering a therapeutically effective amount of an antibiotic to the subject, wherein the subject has been selected as having an increased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject. the use of an antibiotic for the manufacture of a medicament for the treatment of an infection in a subject, wherein the subject has been selected as having an increased quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

A further aspect provides an antibiotic for use in a method of treating an infection in a subject, wherein the method comprises: detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject, such as measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject; and identifying whether the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria is modified, in particular increased.

A related aspect provides a method of treating an infection in a subject in need of such a treatment, the method comprising: detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject, such as measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject; identifying whether the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria is modified, in particular increased; and if the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria is modified, in particular increased, administering a therapeutically effective amount of an antibiotic to the subject. A related aspect provides the use of an antibiotic for the manufacture of a medicament for the treatment of an infection in a subject, wherein the subject has been selected for treatment by a method comprising: detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject, such as measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject; identifying whether the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria is modified, in particular increased; and if the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria is modified, in particular increased, selecting the subject for the treatment.

As used herein, a phrase such as "a subject in need of treatment" includes subjects that would benefit from treatment of a given condition, particularly an infection. Such subjects may include, without limitation, those that have been diagnosed with said condition, those prone to develop said condition and/or those in who said condition is to be prevented.

The terms "treat" or "treatment" encompass both the therapeutic treatment of an already developed disease or condition, such as the therapy of an already developed infection, as well as prophylactic or preventive measures, wherein the aim is to prevent or lessen the chances of incidence of an undesired affliction, such as to prevent occurrence, development, and progression of an infection. Beneficial or desired clinical results may include, without limitation, alleviation of one or more symptoms or one or more biological markers, diminishment of extent of disease, stabilized (i.e., not worsening) state of disease, delay or slowing of disease progression, amelioration or palliation of the disease state, and the like. "Treatment" can also mean prolonging survival as compared to expected survival if not receiving treatment.

The term "prophylactically effective amount" refers to an amount of an active compound or pharmaceutical agent that inhibits or delays in a subject the onset of a disorder as being sought by a researcher, veterinarian, medical doctor, or other clinician.

The methods as taught herein allow to administer a therapeutically effective amount of an antibiotic in subjects having an infection which will not be at risk of developing severe dysbiosis. The term "therapeutically effective amount" as used herein, refers to an amount of active compound or pharmaceutical agent that elicits the biological or medicinal response in a subject that is being sought by a surgeon, researcher, veterinarian, medical doctor, or other clinician, which may include inter alia alleviation of the symptoms of the disease or condition being treated. Methods are known in the art for determining therapeutically effective doses of an antibiotic.

Another aspect of the invention relates to a method for the prevention of severe dysbiosis in a subject, the method comprising administering a vaccine or another product for prevention of infection or isolating the patient in the hospital ward or even deploying advanced cleaning techniques to limit the exposure of the patient to infection, wherein decision to proceed with these procedures is based on the prediction of said subject to be at risk of developing severe dysbiosis by to the methods as taught herein.

Another aspect of the invention relates to a method for the prevention of severe dysbiosis in a subject, the method comprising administering a live biotherapeutic product, such as a probiotic or a fecal microbiota transplant, wherein decision to proceed with these procedures is based on the prediction of said subject to be at risk of developing severe dysbiosis by to the methods as taught herein.

A further aspect relates to a kit of parts comprising a binding agent capable of measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from a subject. A further aspect relates to a kit of parts comprising a set of binding agents capable of measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from a subject. In embodiments, the binding agent may be polynucleotide probe capable of specifically binding to a nucleic acid of the one or more of Faecalibacterium, Casaltella, and/or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria; preferably the binding agent is a set of oligonucleotides capable of specifically binding to a nucleic acid of the one or more of Faecalibacterium, Casaltella, and/or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria.

Preferably, the invention relates to a kit of parts comprising a set of binding agents capable of detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from a subject. In embodiments, the binding agent is polynucleotide probe capable of specifically binding to a nucleic acid of the Faecalibacterium, Casaltella, and Oscillibacter bacteria and a nucleic acid of the one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria; preferably the binding agent is a set of oligonucleotides capable of specifically binding to a nucleic acid of the Faecalibacterium, Casaltella, and Oscillibacter bacteria and a nucleic acid of the one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria.

In embodiments, the kit of parts may comprise a set of oligonucleotides capable of specifically binding to a nucleic acid of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and a nucleic acid of one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria. In embodiments, the kit of parts may comprise a set of oligonucleotides capable of specifically binding to a nucleic acid of the Faecalibacterium, Casaltella, and Oscillibacter bacteria and a nucleic acid of one or more of, such as at least one, at least two, or all three of, Akkermansia, Porphyromonas, or Bifidobacterium bacteria. In embodiments, the kit of parts may comprise a set of oligonucleotides capable of specifically binding to a nucleic acid of the one or more of, such as at least one, at least two, or all three of, Faecalibacterium, Casaltella, or Oscillibacter bacteria and a nucleic acid of Akkermansia, Porphyromonas, and Bifidobacterium bacteria. For instance, the kit of parts may comprise a set of oligonucleotides capable of specifically binding to a nucleic acid of Faecalibacterium bacteria, a nucleic acid of Casaltella bacteria, a nucleic acid of Oscillibacter bacteria, a nucleic acid of Akkermansia bacteria, a nucleic acid of Porphyromonas bacteria, and a nucleic acid of Bifidobacterium bacteria.

The kits of parts may also comprise one or more reference values as defined herein.

The terms "kit of parts" and "kit" as used throughout this specification refer to a product containing components necessary for carrying out the specified methods, packed so as to allow their transport and storage. Materials suitable for packing the components comprised in a kit include crystal, plastic (e.g., polyethylene, polypropylene, polycarbonate), bottles, flasks, vials, ampules, paper, envelopes, or other types of containers, carriers or supports. Where a kit comprises a plurality of components, at least a subset of the components (e.g., two or more of the plurality of components) or all of the components may be physically separated, e.g., comprised in or on separate containers, carriers or supports. The components comprised in a kit may be sufficient or may not be sufficient for carrying out the specified methods, such that external reagents or substances may not be necessary or may be necessary for performing the methods, respectively. Typically, kits are employed in conjunction with standard laboratory equipment, such as liquid handling equipment, environment (e.g., temperature) controlling equipment, analytical instruments, etc. In addition to the binding agents such as the set of isolated oligonucleotides as taught herein, optionally provided on arrays or microarrays, the present kits may also include some or all of solvents, buffers (such as for example but without limitation histidine-buffers, citrate-buffers, succinate-buffers, acetate- buffers, phosphate-buffers, formate buffers, benzoate buffers, TRIS (Tris(hydroxymethyl)- aminomethan) buffers or maleate buffers, or mixtures thereof), enzymes (such as for example but without limitation thermostable DNA polymerase), detectable labels, detection reagents, and control formulations (positive and/or negative), useful in the specified methods. Typically, the kits may also include instructions for use thereof, such as on a printed insert or on a computer readable medium. The terms may be used interchangeably with the term "article of manufacture", which broadly encompasses any man-made tangible structural product, when used in the present context. A further aspect relates to the use of a kit of parts as taught herein for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic, or for predicting severity of dysbiosis after treatment of a subject with an antibiotic. Preferably, the subject is a human subject.

Further aspects relate to the use of a kit of parts as taught herein for selecting a particular antibiotic or class of antibiotics for treatment of a subject; or for guiding treatment of a subject with a particular antibiotic or class of antibiotics. Preferably, the subject is a human subject.

The present application also provides aspects and embodiments as set forth in the following Statements:

Statement 1. A method for predicting severity of dysbiosis after treatment of a subject with an antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

Statement 2. The method according to statement 1, wherein the method comprises:

(b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably wherein the reference value representing a known severity of dysbiosis of a reference subject after treatment with an antibiotic;

(c) finding a deviation or no deviation of the quantity of the bacteria as measured in (a) from said reference value; and (d) attributing the finding of deviation or no deviation to a particular prediction of severity of dysbiosis after treatment of the subject with an antibiotic.

Statement 3. The method according to statement 1 or 1, wherein the method is for predicting severity of dysbiosis after treatment of a subject with a particular antibiotic or a particular class of antibiotics.

Statement 4. A method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

Statement 5. The method according to statement 4, wherein the method comprises:

(b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably the reference value representing a reference subject having severe dysbiosis after treatment with the particular antibiotic or class of antibiotics;

(c) selecting the particular antibiotic or class of antibiotics for treatment of the subject if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is the same or decreased compared with the reference value.

Statement 6. The method according to any one of statements 1 to 5, wherein the method comprises detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota from the subject.

Statement 7. The method according to any one of statements 1 to 6, wherein the method further comprises determining the level of expression of one or more antibiotic resistance genes in a fecal sample or gut microbiota from the subject.

Statement 8. The method according to any one of statements 1 to 7, wherein the detection of the bacteria is performed prior to or during antibiotic treatment; preferably wherein the detection of the bacteria is performed prior to antibiotic treatment. Statement 9. The method according to any one of statements 1 to 8, wherein the reference value is a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria prior to antibiotic treatment of the reference subject.

Statement 10. The method according to any one of statements 1 to 9, wherein: the Faecalibacterium bacteria are Faecalibacterium prausnitzii; the Casaltella bacteria are Casaltella massiliensis; the Oscillibacter bacteria are selected from the group consisting of Oscillibacter massiliensis, Oscillibacter ruminantium, Oscillibacter valericigenes, Oscillibacter avistercoris, Oscillibacter excrementavium, Oscillibacter excrementigallinarum, and Oscillibacter pullicola; the Akkermansia bacteria are Akkermansia muciniphila; the Porphyromonas bacteria are Porphyromonas bennonis; and/or the Bifidobacterium bacteria are selected from the group consisting of Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, and Bifidobacterium catenulatum.

Statement 11. The method according to any one of statements 1 to 10, wherein the step of detecting the bacteria is carried out using nucleic acid sequencing; quantitative polymerase chain reaction (qPCR); reverse transcription polymerase chain reaction (RT-PCR); polymerase chain reaction (PCR); digital PCR; rolling circle amplification (RCA); loop-mediated isothermal amplification (LAMP); a microarray; mass spectrometry; Western blot; immunohistochemistry; enzyme-linked immunosorbent assay (ELISA); or any combination of these methods; and/or wherein the step of comparing the quantity of the bacteria with a reference value may be performed using machine learning, linear discriminant analysis, linear regression, Spearman rank correlation, Euclidean distance; Manhattan distance; Average dot product; Pearson correlation; Pearson uncentered; Pearson squared; Cosine correlation; Covariance value; Kendall's Tau; or Mutual information.

Statement 12. The method according to any one of statements 1 to 11, wherein the fecal sample is a rectal swab or a stool sample.

Statement 13. The method according to any one of statements 1 to 12, wherein the antibiotic or class of antibiotics is selected in group consisting of beta-lactams, beta-lactam and beta-lactamase inhibitor combinations, penicillins, penicillin and beta-lactamase inhibitor combinations, penicillinase-resistant penicillins, penicillinase-resistant penicillin and beta- lactamase inhibitor combinations, cephalosporins, cephalosporin and beta-lactamase inhibitor combinations, carbapenems, carbapenem and beta-lactamase inhibitor combinations, monobactams, quinolones, fluoroquinolones, sulfonamides, aminoglycosides, tetracyclines, macrolides, glycopeptides, oxazolidinones, phenicols, lincosamides, Streptogramins, polymyxins, diaminopyrimidines, sulfones, para-aminobenzoic acid, bacitracin, isoniazid, rifamycins ethambutol, ethionamide, capreomycin, and clofazimine; the class of antibiotics is a penicillin in combination with a beta-lactamase inhibitor, a betalactam antibiotic in combination with a beta-lactamase inhibitor, a beta-lactam antibiotic, or a fluoroquinolone antibiotic; and/or the antibiotic is ampicillin in combination with sulbactam; amoxicillin in combination with clavulanic acid; piperacillin in combination with tazobactam; ceftriaxone; meropenem; ciprofloxacin; or levofloxacin.

Statement 14. A kit of parts comprising a binding agent capable of detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota from a subject.

Statement 15. The kit of parts according to statement 14, wherein the binding agent is polynucleotide probe capable of specifically binding to a nucleic acid of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria; preferably the binding agent is a set of oligonucleotides capable of specifically binding to a nucleic acid of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria.

The herein disclosed aspects and embodiments of the invention are further supported by the following non-limiting examples.

EXAMPLES

Example 1: Methods according to embodiments of the present invention for predicting whether a subject is at risk of developing severe dysbiosis after treatment with an antibiotic and for selecting a particular class of antibiotics or antibiotic for treatment of a subject

This study aimed to characterize microbial perturbations as a result of treatment with seven individual broad-spectrum antibiotics: ampicillin/sulbactam, amoxicillin/clavulanic acid, piperacillin/tazobactam, ceftriaxone, meropenem, ciprofloxacin, and levofloxacin by utilizing 16S rRNA gene profiling of longitudinally collected fecal samples from a diverse, elderly, hospitalized, European population. Furthermore, we aimed to define a core set of bacterial taxa sensitive to antibiotic treatment that could be utilized to predict post-treatment microbial dysbiosis linked to the risk of adverse events.

Materials and methods

Study design

The samples used in this study were collected as part of the ANTICIPATE trial (ClinicalTrials.gov NCT02896244), a multi-center prospective observational study including 1,007 patients conducted in 34 European hospitals in Germany, Greece, France, Romania, Spain, and the Netherlands. Inclusion criteria included hospitalization, an age of 50 years or above, and treatment with one or several of the broad-spectrum antibiotics: penicillins/beta-lactamase inhibitors, carbapenems, 3^rd or 4^th generation cephalosporins, fluoroquinolones, or clindamycin. Two rectal swab samples were collected from each patient: one at the start of administration of antibiotics (DI) and one six days later (D6) ± 24 h or at hospital discharge if within the first six days of the study period.

Analysis population

In this study, patients receiving treatment with a single antibiotic between the first (DI) and second (D6) fecal sample collection were considered for analysis (n = 310). Antibiotic courses administered on D6, i.e., the day of the second sample collection, were not considered to impact the microbial flora within the studied timeframe and were therefore disregarded. Antibiotic classification was defined according to the 2019 ATC index (WHO Collaborating Centre for Drug Statistics Methodology), based on their chemical composition and mechanism of action. The effect of the seven broad-spectrum antibiotics ampicillin/sulbactam (SAM), amoxicillin/clavulanic acid (AMC), piperacillin/tazobactam (TZP), ceftriaxone (CRO), meropenem (MEM), ciprofloxacin (CIP), and levofloxacin (LVX) were investigated. Scarce groups with less than 10 patients receiving a single antibiotic were excluded from further analysis.

Sample collection, DNA isolation, library preparation, and sequencing

Fecal samples collected at DI and D6 (n = 310 each) were characterized using 16S rRNA gene profiling as described in the ANTICIPATE study (van Werkhoven et al., 2021, Nat. Commun., 12, 2240; Berkell et al. 2021, Nat. Commun., 12, 2241). Briefly, total DNA was extracted using the FastDNA SPIN Kit (MP Biomedicals, Santa Ana, USA) according to the manufacturer's instructions where DNA quality was assessed by agarose gel electrophoresis and the concentration measured with the Qubit ds DNA HS Assay Kit in a Qubit 3.0 Fluorometer (ThermoFisher Scientific, Waltham, USA). Extracted DNA was used as template for PCR amplification of the V3-V4 regions of the 16S rRNA gene followed by Nextera XT library preparation and 2 x 250 bp or 2 x 300 bp paired-end sequencing on the MiSeq platform (Illumina Inc., San Diego, USA) as described by the manufacturer. The samples were sequenced together with positive mock community controls (HM-783D, https://www.beiresources.org/), sample controls as well as negative controls such as negative PCR controls and negative DNA extraction controls to verify study reproducibility and accuracy, as previously described by Berkell et al. (supra).

16S rRNA gene profiling

Pre-processing, quality filtering, and classification

Pre-processing of raw sequence data was conducted using the OCToPUS pipeline vl.O implementing SPAdes V3.5.0, IPED vl.O, CATCh vl.O, mothur vl.39.1, and UPARSE (USEARCH v8.1.186 implementation) for cleaning, denoising, chimera removal, and clustering of the reads. Samples were rarefied to 15,000 reads for all subsequent analysis to avoid bias, as described previously by Berkell et al. (supra). Alpha and beta diversity indices Shannon, Chaol, and weighted UniFrac were calculated in mothur. Sequenced samples were de novo-clustered into microbial community types (MCTs) using the mothur-implementation of the Dirichlet Multinomial Mixtures (DMM) algorithm. Single-nucleotide resolution of reads clustered within differential abundant OTUs were subjected to oligotyping to delineate OTUs into species level when possible. Identified oligotypes with a total abundance > 1% within the OTU were classified using NCBI nucleotide blast against the 16S rRNA database where hits with >97% identity were considered true.

Statistical analysis and visualization

Cross-sectional and longitudinal comparisons of alpha diversity indices was conducted using two- sided, non-parametric Friedman rank sum (FRS) or Kruskal-Wallis (KW) tests with Mann-Whitney (MW) or Wilcoxon signed rank (WSR) post-hoc tests followed by Bonferroni correction of p-values as indicated for paired and non-paired comparisons, respectively. Beta diversity was compared between patient groups using analysis of molecular variance (AMOVA).

Differential abundant OTUs were identified using Linear Discriminant Analysis Effect Size (LEfSe) for cross-sectional comparisons or by first assessing normality and homoscedasticity using Shapiro's and Bartlett's tests, respectively, followed by paired WSR or t-tests as indicated with Benjamini- Hochberg FDR p-value correction for longitudinal comparisons. Only OTUs with a median relative abundance > 0.05% and either an LDA > 3.0 (cross-sectional) or a fold change > 3.0 (longitudinal) were reported. These differential abundant OTUs were used to calculate a microbial dysbiotic index (MDI) defined as the ratio between the total abundance of OTUs with increasing abundance after antibiotic treatment (at D6) and the total abundance of OTUs with decreasing abundance after antibiotic treatment (at D6) as described by Gevers et. al., 2014, Cell Host Microbe, 15, 382-392. Microbiota-associated metabolic function was inferred using PICRUSt2 v.2.1.3 with KEGG Orthology (KOs) reactions from the IMG v.4 database as reference using Welch's non-parametric t-test reporting pathways with an effect size > 1.5. P-values < 0.05 were considered significant for all statistical analyses in this study.

Statistical analyses and data visualization was conducted in RStudio v.3.5.0 using R v.4.0.2 (https://www.R-project.org/) with the tidyversev 1.3.0 package, STAMP v.2.1.3, or using Circos, and images were annotated in Inkscape v.1.0.

Microbiota-based classifiers for prediction of dysbiosis post antibiotic treatment

Differential abundant OTUs identified by LEfSe (LDA > 3.0) when comparing MCT1- and MCT2- patients at DI (Table 3) were utilized for prediction of MCT2 communities post treatment.

These individual OTUs together with their summation, country of origin, age, gender, alpha diversity, Charlson comorbidity index, and Karnofsky score as well as antibiotic compounds consumed, dose, duration, administration route were used to train machine learning classifiers using WEKA v3.7.11. The data were first filtered by selecting the most informative subset of the OTUs and their summation using the AttributeSelectedClassifier model that applies correlationbased feature selection. This approach evaluates the utility of a subset of the OTUs by considering the individual predictive ability of each OTU along with the degree of redundancy between them. An artificial neural network classifier was then used to classify each patient into MCT1 and MCT2, using the MultilayerPerceptron classifier. A leave-one-out cross validation approach was used to separate the data into a training set and an unexposed test set by segregating the data into partitions, and repeating this partitioning iteratively and by using a different partition for testing each time. To account for the unbalanced ratio of MCT1 to MCT2, the CostSensitiveClassifier function was applied, which performs reweighting of training features according to the cost assigned to each misclassification.

Similarly, class-specific models were developed stratifying patients into groups based on antibiotic classes received. These groups included patients receiving penicillin/beta-lactamase inhibitor combinations (SAM, AMC, TZP), other beta-lactam antibiotics (CRO, MEM), and fluoroquinolones (CIP, LVX). Stratification according to individual antibiotics compounds was also performed but involved few observations per group, specifically for MCT2-classified patients.

Three parameters were adopted for evaluation of the trained classifiers; sensitivity [TP/(TP + FN )], specificity [TN/(TN + FP)], and the area under the curve (AUC) for the receiver operating characteristic (ROC) analysis that combines both sensitivity and specificity into one parameter were defined by TP (correctly classified MCT2), FN (MCT2 classified as MCT1), TN (correctly classified

MCT1), and FP (MCT1 classified as MCT2).

Results

Study population and design

This study utilized a sub-selection of samples collected in the prospective, observational, cohort study ANTICIPATE ("AssessmeNT of the Incidence of Clostridium difficile Infections in hospitalized Patients on Antibiotic TrEatment", ClinTrial.gov: NCT02896244). ANTICIPATE recruited a total of 1,007 patients receiving broad-spectrum antibiotic treatment where rectal swab samples were collected from 1,002 patients at study enrolment (DI) and an additional 848 were collected approximately six days later (D6). A total of 310 patients who provided samples before (DI) and after treatment (D6) with a single broad-spectrum antibiotic were selected and further analysed for assessment of antibiotic-specific dysbiosis and impact on associated metabolic function of the intestinal microbiota in this study (Figure 1).

The analysed patient population in this study had a median age of 71 years (IQR: 63-80), a Charlson comorbidity index of 5.2 (IQR: 3.8-6.8), a Karnofsky score of 80 (IQR: 70-90), and 57.4% of the study subjects were male (n = 178). Among antibiotics, beta-lactams were most commonly prescribed; 160 patients (51.6%) were treated with penicillins combined with beta-lactamase inhibitors, where 23, 98, and 39 patients received ampicillin/sulbactam (SAM), amoxicillin/clavulanic acid (AMC), and piperacillin/tazobactam (TZP), respectively; 97 patients (31.3%) received other beta-lactams, where 79 patients were treated with the 3^rd generation cephalosporin, ceftriaxone (CRO), and 18 with the carbapenem, meropenem (MEM). Finally, 63 patients (20.3%) received treatment with fluoroquinolones where 24 patients received ciprofloxacin (CIP) and 29 received levofloxacin (LVX).

Unsupervised clustering reveals moderately or severely dysbiotic microbial communities after antibiotic treatment

To identify community types characterized by large differences in microbial composition, unsupervised, de novo clustering of samples collected at pre- (DI) and post-antibiotic (D6) timepoints was performed separately. While no distinct segregation could be reported at DI, samples collected at D6 clustered into two distinct microbial community types (MCTs), MCT1 and MCT2, that comprised 74.2% (n = 230) and 25.8% (n = 80) of the D6 samples, respectively.

When stratified by antibiotic compound, it was found that MEM-, CIP-, and TZP-treated patients were overrepresented in the MCT2 group at D6 (72.2%, 45.8%, and 30.8% of treated patients, respectively), whereas SAM-, AMC-, and LVX-treated patients were underrepresented (13.0%, 16.3%, and 13.8% of treated patients, respectively, Table 1). Table 1: Patient distribution between MCT1 (n = 230) and MCT2 (n = 80) clusters at D6.

Patients stratified by MCT display unique microbial pre-treatment composition

Patient microbiota at baseline was primarily characterized by elevated levels of members within the Lachnospiraceae family when compared to D6-MCT1 and D6-MCT2 communities (Table 2).

Common to both baseline and D6-MCT1-Iike communities were elevated levels of Anaerococcus, Bifidobacterium, Blautia, Campylobacter, Peptoniphilus, and Prevotella spp. compared to D6-MCT2. D6-MCT1 communities were more characterized by Ruminococcaceae members compared to baseline (Table 2, Figure 2). The more discrete cluster, D6-MCT2, harbored elevated levels of Bacteroides spp. as well as potential enteropathogens such as Enterococcus and Escherichia/Shigella spp (Table 2, Figure 2).

Table 2: Cross-sectional comparison of microbial composition between patients at baseline (DI) and after treatment based on identified microbial clusters (MCT1 and MCT2 at D6). Differences in microbial composition was assessed using LEfSe (LDA > 3.0) for samples clustered together; baseline (DI, n = 310), MCT1 (D6, n = 230), and MCT2 (D6, n = 80). DI: rectal swab sample collected upon study enrolment; D6: rectal swab collected approximately six days after initiation and at the end of antibiotic treatment. MCT: microbial community type. LEfSe: linear discriminant analysis effect size. LDA: Linear discriminant analysis score. OTU: Operational taxonomic unit.

Although unsupervised clustering could not segregate samples collected at DI into distinct microbial communities, statistical assessment of microbial diversity and composition revealed that patients who developed MCT1 and MCT2 profiles post treatment were found to harbor higher Shannon diversity and Chaol richness (Mann-Whitney, p < 0.003) as well as a distinct microbial composition at DI (AMOVA, p = 0.023). Further, pre-treatment (DI) microbial profiles for MCT1 and MCT2 patients showed that MCTl-classified patients harbored slightly lower MDI values compared to MCT2 patients at DI (MCT1: median -0.765 [IQR: -1.32 - -0.141; MCT2: median -0.319 [IQR: - 1.18 - 0.329]; Mann-Whitney: p = 0.021), indicating that pre-treatment differences could be attributable to differences in baseline microbial composition. When comparing taxonomical differences in microbial composition between MCT1 and MCT2 communities, it was found that D1-MCT1 communities were characterized by elevated levels of Faecalibacterium, Akkermansia, Porphyromonas, Oscillibacter, and Bifidobacterium spp. as well as uncultured members of the Clostridiales order (Figure 2, Table 3). In contrast, these taxa were largely absent in D1-MCT2 communities.

Table 3: Cross-sectional comparison of microbial composition between patients classified as MCT1 and MCT2 at DI. Differences in microbial composition was assessed using LEfSe (LDA > 3.0) between patients classified MCT1 (n = 230) and MCT2 (n = 80) at DI. MCT: microbial community type. LEfSe: linear discriminant analysis effect size. LDA: Linear discriminant analysis score. OTU: Operational taxonomic unit.

Oligotyping further delineation these taxa into the following species: Faecalibacterium prausnitzii, Akkermansia muciniphila, Porphyromonas bennonis, Oscillibacter spp., Casaltella massiliensis, and Bifidobacterium spp. (B. adolescentis, B. faecale, B. stercoris, B. pseudocatenulatum, B. catenulatum) (Table 4). Only oligotypes with a total abundance > 1% were classified. In case multiple oligotypes were identified with the same classification these were merged in the table displaying a range of oligotypes and NCBI blast identities. Only hits with an NCBI blast identity > 97% was considered valid.

Table 4: Oligotyping of the operational taxonomical units to identify individual species.

When combining baseline characteristics such as age, gender, country of origin, and comorbidity score along with alpha and beta diversity, abundances of MCT-specific biomarkers at DI and D6, antibiotic treatment, and the 90-day diarrheal incidence, we clearly observe that MCT2 patients harbor a severely dysbiotic microbiota post antibiotic treatment with higher frequency of adverse events whereas MCT1 patients develop a mild dysbiosis with intermediate microbial diversity. Stratification of MCT-patients by antibiotic treatment, showed that MCT2-patients suffer a greater reduction in alpha diversity post treatment irrespective of antibiotic compound. Baseline characteristics, except the differences in microbial composition at DI, were comparable between the two MCT groups. Collectively, these findings support the hypothesis that the post-treatment MCT is largely determined by the presence or depletion of the Dl-specific biomarkers, namely the level of bacteria of Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium (Table 3).

Machine learning classification enables prediction of post-antibiotic MCT and severity of dysbiosis

To predict post-antibiotic MCT based on the DI microbial composition, we constructed a dataset with OTUs predicted by LEfSe, their summation and the antibiotic compound used for treatment. The latter two were detected to be the most informative features, and thus utilized for the training. By applying an artificial neural network classifier with cross validation utilizing the identified differential abundant OTUs (Table 3), it was possible to construct a generalized model for the prediction of MCT2 communities post-treatment. This resulted in the ability to predict MCT2-like communities post antibiotic treatment with a sensitivity and specificity of 64% and 62%, respectively (Table 5).

Table 5: Method according to an embodiment illustrating that machine learning classifiers can predict post-antibiotic dysbiosis. Cl: confidence interval. MCT: microbial community type.

Machine learning classification enables selection of class of antibiotic or antibiotic for treatment

Class-specific classifiers for penicillin/beta-lactamase inhibitor combinations (SAM, AMC, TZP), other beta-lactam antibiotics (CRO, MEM), and fluoroquinolones (CIP, LVX) were constructed by combining observations for antibiotics of the same class. Thus, it was possible to construct class- specific artificial neural network classifiers predictive of MCT2-like community development posttreatment. For penicillin/beta-lactamase inhibitor combinations, other beta lactam antibiotics, and fluoroquinolones the sensitivities of these models were 68%, 62%, and 67%, respectively, whereas the specificities where 61%, 64%, and 61%, respectively (Table 5).

Further, antibiotic-specific classifiers for individual antibiotics: SAM, AMC, TZP, CRO, MEM, CIP, and LVX were constructed. It was possible to construct antibiotic-specific artificial neural network classifiers predictive of MCT2-like community development post-treatment but as the number of MCT2 observations were low for several individual antibiotics, sensitivity and specificity were lower for some of the antibiotics (Table 5).

Claims

1. A method for predicting prior to an envisaged treatment of a subject with an antibiotic whether the subject is at risk of developing severe dysbiosis after treatment with the antibiotic, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject.

2. The method according to claim 1, wherein the method comprises:

(a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject;

(b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably wherein the reference value represents a reference subject having developed severe dysbiosis after treatment with an antibiotic; and

(c) predicting that the subject is at risk of developing severe dysbiosis after treatment with an antibiotic if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is the same or decreased compared with the reference value.

3. The method according to claim 1 or 2, wherein the method is for predicting whether a subject is at risk of developing severe dysbiosis after treatment of the subject with a particular antibiotic or a particular class of antibiotics.

4. A method for selecting a particular antibiotic or class of antibiotics for treatment of a subject, the method comprising detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject prior to antibiotic treatment.

5. The method according to claim 4, wherein the method comprises:

(a) measuring the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in the fecal sample or gut microbiota sample obtained from the subject prior to antibiotic treatment; (b) comparing the quantity of the bacteria as measured in (a) with a reference value of the quantity of one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria, preferably the reference value representing a reference subject having severe dysbiosis after treatment with the particular antibiotic or class of antibiotics;

(c) selecting the particular antibiotic or class of antibiotics for treatment of the subject if the quantity of the one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria as measured in (a) is changed compared with the reference value.

6. The method according to any one of claims 1 to 5, wherein the method comprises detecting Faecalibacterium, Casaltella, and Oscillibacter bacteria in addition to one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject.

7. The method according to any one of claims 1 to 6, wherein the method comprises detecting Faecalibacterium, Casaltella, Oscillibacter, Akkermansia, Porphyromonas, and Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from the subject.

8. The method according to any one of claims 1 to 7 , wherein the method further comprises determining the level of expression of one or more antibiotic resistance genes in a fecal sample or gut microbiota sample obtained from the subject.

9. The method according to any one of claims 1 to 8, wherein the reference value is a reference value of the quantity of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria prior to antibiotic treatment of the reference subject.

10. The method according to any one of claims 1 to 9, wherein: the Faecalibacterium bacteria are Faecalibacterium prausnitzii; the Casaltella bacteria are Casaltella massiliensis; the Oscillibacter bacteria are selected from the group consisting of Oscillibacter massiliensis, Oscillibacter ruminantium, Oscillibacter valericigenes, Oscillibacter avistercoris, Oscillibacter excrementavium, Oscillibacter excrementigallinarum, and Oscillibacter pullicola; the Akkermansia bacteria are Akkermansia muciniphila; the Porphyromonas bacteria are Porphyromonas bennonis; and/or the Bifidobacterium bacteria are selected from the group consisting of Bifidobacterium adolescentis, Bifidobacterium faecale, Bifidobacterium stercoris, Bifidobacterium pseudocatenulatum, and Bifidobacterium catenulatum.

11. The method according to any one of claims 1 to 10, wherein the step of detecting the bacteria is carried out using nucleic acid sequencing; quantitative polymerase chain reaction (qPCR); reverse transcription polymerase chain reaction (RT-PCR); polymerase chain reaction (PCR); digital PCR; rolling circle amplification (RCA); loop-mediated isothermal amplification (LAMP); a microarray; mass spectrometry; Western blot; immunohistochemistry; enzyme-linked immunosorbent assay (ELISA); or any combination of these methods; and/or wherein the step of comparing the quantity of the bacteria with a reference value may be performed using machine learning, linear discriminant analysis, linear regression, Spearman rank correlation, Euclidean distance; Manhattan distance; Average dot product; Pearson correlation; Pearson uncentered; Pearson squared; Cosine correlation; Covariance value; Kendall's Tau; or Mutual information.

12. The method according to any one of claims 1 to 11, wherein the fecal sample is a rectal swab or a stool sample.

13. The method according to any one of claims 1 to 12, wherein the antibiotic or class of antibiotics is selected in group consisting of beta-lactams, betalactam and beta-lactamase inhibitor combinations, penicillins, penicillin and betalactamase inhibitor combinations, penicillinase-resistant penicillins, penicillinaseresistant penicillin and beta-lactamase inhibitor combinations, cephalosporins, cephalosporin and beta-lactamase inhibitor combinations, carbapenems, carbapenem and beta-lactamase inhibitor combinations, monobactams, quinolones, fluoroquinolones, sulfonamides, aminoglycosides, tetracyclines, macrolides, glycopeptides, oxazolidinones, phenicols, lincosamides, Streptogramins, polymyxins, diaminopyrimidines, sulfones, paraaminobenzoic acid, bacitracin, isoniazid, rifamycins ethambutol, ethionamide, capreomycin, and clofazimine; the class of antibiotics is a penicillin in combination with a beta-lactamase inhibitor, a beta-lactam antibiotic in combination with a beta-lactamase inhibitor, a beta-lactam antibiotic, or a fluoroquinolone antibiotic; and/or the antibiotic is ampicillin in combination with sulbactam; amoxicillin in combination with clavulanic acid; piperacillin in combination with tazobactam; ceftriaxone; meropenem; ciprofloxacin; or levofloxacin.

14. A kit of parts comprising one or more binding agents capable of detecting one or more of Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria in a fecal sample or gut microbiota sample obtained from a subject.

15. The kit of parts according to claim 14, wherein the binding agents comprise one or more polynucleotide probes capable of specifically binding to a nucleic acid of the Faecalibacterium, Casaltella, or Oscillibacter bacteria and one or more polynucleotide probes capable of specifically binding to a nucleic acid of one or more of Akkermansia, Porphyromonas, or Bifidobacterium bacteria; preferably the binding agent is a set of oligonucleotides capable of specifically binding to a nucleic acid of the Faecalibacterium, Casaltella, Oscillibacter Akkermansia, Porphyromonas, and Bifidobacterium bacteria.