faecal microbiota transplants: the evidence and...

Faecal Microbiota Transplants: The evidence and experience Dr Simon Goldenberg Consultant Microbiologist and Infection Control Doctor Guy’s & St Thomas’ NHS Foundation Trust

C.difficile

ESBL/VRE/postantibioticdysbiosis

IBDandIBS

Insulinresistance/type2diabetes/obesity

Auoimmunediseasese.g.coeliac

Gut-brainaxisdiseasese.g.autism,MS,chronicfatiguesyndrome,parkinsons

Adapted from:Aviv Cohen. Dig Dis Sci 2016 62:5, 1131-1145Bakker Microbiol Spectr 2017 75:4

Gut microbiota and health

Causality Level of complexity Course of disease Responders / non-responders Influence of genetics, diet Concomitant medication

The burden of CDI recurrence

20%

40%

60%

After first episode After first recurrence After two or more recurrences

KellyCP,LaMontJT.NEnglJMed.2008;359(18):1932-1940;JohnsonS.InternationalJournalofAntimicrobialAgents33(2009)S33-S36.Deshpandeetal.ICHE201536:4,452-60

–  Age≥65years1-4,10–  SevereCDI1,4–  PreviousepisodeofCDI1,3-5–  Co-morbiditiesincludingrenalfailure1,4,6,7,10–  Exposuretoconcomitantantibiotics4,10–  InfectionwithparticularstrainsegRT027,RT106–  ExposuretoPPIs/othergastricacidsupressors8,10

–  Previoushospitaladmission9

1Kyneetal.Lancet2001;357:189-932Baueretal.ClinMicrobiolInfect2009;15:1067-793Baueretal.Lancet2011;377:63-734Huetal.Gastro2009;136:1206-145McFarlandetal.AmJGastroenterol2002;97:1769-756Doetal.CID1998;26:954-97Baueretal.ClinMicrobiolInfect2011;17:A1-48Kwoketal.AmJGastro2012;107:1011-99Eyreetal.CID2012;55:77-8810Deshpandeetal.ICHE201536:4,452-60

Decreased Diversity of the Faecal Microbiome in Recurrent C. difficile

PatientswithrecurrentC.difficilehavedecreasedphylogeneticrichness

ChangJY,etal.JInfectDis2008:197;435-8

SomebacteriaremaindisruptedforprolongedperiodsfollowingantibioticadministrationSadowskyetal.TheFecalBacteria,2011Zauraetal.mBio6:6Nov/Dec2015

Faecal Microbiota Transplantation (FMT)

Rationale:Aperturbedimbalanceinintestinalmicrobiota(dysbiosis)isassociatedwithorcausesdiseaseandcanbecorrectedbyre-introductionofdonorfaeces

Normalmicrobiota

Competitivenicheexclusion

DirectinhibitionofC.difficilegrowthanditstoxigenicactivity

Modulationofmetabolites–bileacidtransformation

Triggerofimmunemediatedresponses

Mechanisms by which the effects of a restored microbiota prevent CDI recurrence

4thcentury:GeHongdescribeduseofhumanfaecalsuspensionbymouthfordiarrhoeaandfoodpoisoning“ZghouHouBeiJiFang”HandyTherapyforEmergencies

Early History of FMT

16thcentury:LiShizhenprescribedfermentedfaecesforabdominaldiseaseswithdiarrhoea,abdominalpain,fever,vomitingandconstipation;“yellowdragonsoup”“BenCaoGangMu”CompendiumofMateriaMedica

Early History of FMT

17thcentury:veterinarymedicine:

Transfaunation(transferofcecalcontentsorfreshfaeces)fromhealthyhorsestotreathorseswithchronicdiarrhoeaRumentransfaunationisusedto‘refaunate’cowsthathavebeenoff-feedbecauseofillnesse.gmastitis

Modernhistory:(1958)Eismannetal.4patientswithfulminantpseudomembranouscolitistreatedwithFMTenema

Studyandpatientcharacteristics Outcomes Adverseevents

Intervention=FMT(nasoduodenal)+vanc+bowellavageN=16,F:M=8Age(mean/median)=73+/-13Comparator=vancN=13,F:M=7Age(mean/median)=66+/-14Comparator=vanc+bowellavageN=13,F:M=3Age(mean/median)=69+/-16Pre-FMTantibiotics=Atleastonecourseofadequateantibiotictherapy(>=10daysofVanc>=125mgqdsor>=10daysofMetronidazole500mgtds).Totalfollowupperiod=Afterfirstinfusionat10weeks;followupwasextendedto10weeksafter2ndinfusionRiskofbiasassessment=LOW

Treatmentarm=FMT+vanc+bowellavageOverallcurerate=15/16(94%)*Curewith1infusionalone=13/16(81%)*Treatmentarm=vancOverallcurerate=4/13(31%)*patientsat10weeksTreatmentarm=vanc+bowellavageOverallcurerate=3/13(23%)*patientsat10weeks*P


Intervention=FMT(colonoscopy)N=20,F:M=12:8Age(mean/median)=71(range29-89)Comparator=VancomycinN=19,F:M=11:8Age(mean/median)=75(range49-93)Pre-FMTantibiotics=Vanctaper19+/-1metronidazoleTotalfollowupperiod=10weeksRiskofbiasassessment=UNCERTAINriskofbias

Treatmentarm=FMTOverallcurerate=18/20(90%)*Curewith1infusionalone=13/20(65%)Treatmentarm=VancomycinOverallcurerate=5/19(26%)

MinorGIadverseevents=19diarrhoea,12bloating(allresolvedat12hrs)Minornon-GIadverseevents=Seriousadverseevents=noneDeaths=2-fromC.diffrelatedcomplications.

Cammarota et al, Alimentary Pharmacol & Therap, 2015


Intervention=FMT(enema)N=16,F:M=11Age(mean/median)=Mean75.7Comparator=6weekvanctaperN=12,F:M=8Age(mean/median)=Mean69.6Pre-FMTantibiotics=Atleast1courseofvancomycinformin10daysTotalfollowupperiod=120daysRiskofbiasassessment=UNCERTAINriskofbias

Treatmentarm=FMTOverallcurerate=7/16(44%)Treatmentarm=6weekvanctaperOverallcurerate=7/12(58%)P=ns,discontinuedfollowingfutilityanalysis

MinorGIadverseevents=abdominalpain,tendernessandbloating,equalinbothgroupsMinornon-GIadverseevents=Seriousadverseevents=1developedanasarcafromliverdisease,1hadperfbowelfromdiverticulitis35dayspostFMTDeaths=None.

Hota et al, Clinical Infectious Diseases, 2016


Intervention=FrozenFMT(enema)N=108,F:M=72Age(mean/median)=Meanage73yearsComparator=FreshFMT(enema)N=111,F:M=74Age(mean/median)=Meanage73yearsPre-FMTantibiotics=MetronidazoleandVancomycinTotalfollowupperiod=13weeksRiskofbiasassessment=LOWriskofbias

Treatmentarm=FrozenOverallcurerate=98/109(90.7%)Curewith1infusionalone=57/108(52.8%)Treatmentarm=FreshOverallcurerate=95/111(85.6%)Curewith1infusionalone=56/111(50.5%)

MinorGIadverseevents=Transientdiarrhoea(70%),abdominalcramps(10%),nausea(5%)in24hourspostFMT;constipation(20%)andflatulence(25%)infollowupperiod.NodifferencebetweenthetwogroupsMinornon-GIadverseevents=Seriousadverseevents=12patientsrequiredhospitalizationunrelatedtoFMTDeaths=6deathsinfrozenand13deathsinfresharm(unrelatedtoFMT).

Lee et al, JAMA, 2016


Intervention=Fresh(colonoscopy)N=25,F:M=21Age(mean/median)=Mean75(19-97)Comparator=Lyophilised(colonoscopy)N=23,F:M=13Age(mean/median)=63(20-87)Comparator=Frozen(colonoscopy)N=24,F:M=18Age(mean/median)=62.5(33-88)Pre-FMTantibiotics=NotstatedTotalfollowupperiod=2monthsRiskofbiasassessment=HIGHriskofbias

Treatmentarm=FreshOverallcurerate=25/25(100%)Treatmentarm=FrozenOverallcurerate=20/24(83%)Treatmentarm=LyophilisedOverallcurerate=18/23(78%)Freshvs.lyophilisedP=0.022

MinorGIadverseevents=nodifferencesinthethreegroups.mildtransientabdopainanddiarrhoeain86%ofpatients.6experiencesfatigueand4hadaheadache.2gainedweightMinornon-GIadverseevents=Seriousadverseevents=noneDeaths=none.

Jiang et al, Alimentary Pharmacol & Therap, 2017


Intervention=OverallN=43,F:M=30Age(mean/median)=NotstatedComparator=ColonoscopyN=21,F:M=13Age(mean/median)=Mean67.8Comparator=CapsuleN=22,F:M=17Age(mean/median)=Mean66.3Pre-FMTantibiotics=notreportedTotalfollowupperiod=NotreportedRiskofbiasassessment=UNCERTAINriskofbias

Treatmentarm=OverallOverallcurerate=41/43(95%)Treatmentarm=ColonoscopyOverallcurerate=20/22(91%)Treatmentarm=CapsuleOverallcurerate=21/21(100%)

MinorGIadverseevents=Transientnauseaandvomitingin2overallMinornon-GIadverseevents=Seriousadverseevents=NilDeaths=Nil.

Kao et al, Canadian J Gastroenterol & Hepatol (abstract), 2016


Intervention=DonorFMT(colonoscopy)N=22,F:M=18Age(mean/median)=Meanage48(SD16)Comparator=AutologousFMT(colonoscopy)N=24,F:M=19Age(mean/median)=Meanage55(SD14)Totalfollowupperiod=8weekoutcomefollowup,6monthsafetyfollowupRiskofbiasassessment=LOWriskofbias

Treatmentarm=DonorFMTOverallcurerate=20/22(90.9%)Treatmentarm=AutologousFMTOverallcurerate=15/24(62.5%)P=0.042

MinorGIadverseevents=RatesofothersolicitedAEs(fever,abdominalpain,bloating,nausea,vomiting,diarrhea,flatulence,anorexia,andconstipation)didnotdiffersignificantlybetweengroups.Minornon-GIadverseevents=Seriousadverseevents=NonedescribedDeaths=None.

Kelly et al, Annals of Internal Medicine, 2016

Studyandpatientcharacteristics Outcomes AdverseeventsIntervention=ColonoscopyN=10,F:M=6Age(mean/median)=Mean50.4Intervention=NGN=10,F:M=5Age(mean/median)=Mean58.6Pre-FMTantibiotics:treatmentfailuresofa6-to8-weektaperwithVancomycin(95%ofpatients)withorwithoutanalternativeantibiotic,includingFidaxomicin(70%ofparticipants).Totalfollowupperiod=8weeksfollowupforprimaryresponseRiskofbiasassessment=UNCERTAINriskofbias

Treatmentarm=OverallOverallcurerate=18/20(90%)Curewith1infusionalone=14/20(70%)Treatmentarm=ColonoscopyOverallcurerate=10/10(100%)Curewith1infusionalone=8/10(80%)Treatmentarm=NGOverallcurerate=8/10(80%)Curewith1infusionalone=6/10(60%)P=ns

MinorGIadverseevents=Mildabdominaldiscomfortandbloatingin4patients(20%).Onechildtreatedcolonoscopicallyhadatransientfeverof38.8Conday2thatresolvedspontaneouslyMinornon-GIadverseevents=Seriousadverseevents=1newdiagnosisofmalignancy,1hospitalisationforFourniergangreneDeaths=2deaths(unrelated).

Youngster et al, Clinical infectious diseases, 2014


Intervention=LowdoseFMTcapsules(x30once)N=10,F:M=nsAge(mean/median)=nsComparator=HighdoseFMTcapsules(x30dailyontwoconsecutivedays)N=9,F:M=nsAge(mean/median)=nsPre-FMTantibiotics=nsTotalfollowupperiod=8weeksRiskofbiasassessment=UNCERTAINriskofbias

Treatmentarm=LowdoseFMTcapsules(30pillsonce)Overallcurerate=7/10(70%)Treatmentarm=HighdoseFMTcapsules(30pillsdailyontwoconsecutivedays)Overallcurerate=7/9(78%)

MinorGIadverseevents=noneMinornon-GIadverseevents=Seriousadverseevents=noneDeaths=none.

Allegretti et al, Gastroenterology (DDW abstract), 2016

lower GI-delivered FMT was 95% (95% CI 92%-97%) with likely

moderate heterogeneity between the studies (I2=48%). When

adjusted for funnel plot asymmetry overall response was 90%.

This compared to the overall pooled response rate to upper GI

FMT of 88% (95% CI 82%-94%) with moderate observed hetero-

geneity between the studies adjusting for funnel plot asymmetry

revealing a response rate of 83%. There was evidence of a differ-

ence between the delivery methods with respect to response to

FMT (P=.02). Analysis of cure rate with a single infusion did not

show a significant difference with route of delivery with a

response of CDI of 81% (95% CI 73%-88%) to lower GI-delivered

FMT and of 87% (95% CI 79%-94%) for upper GI-delivered FMT

(P=.20).

3.3.4 | Comparison of freshly prepared vs frozenFMT

In the RCT by Lee, patients received either FMT prepared no

more than 5 hours earlier (n=111) or FMT frozen for up to

30 days (n=108).24 The clinical resolution of diarrhoea using

intention-to-treat analysis showed no evidence of a difference in

outcome between preparations, with a relative risk of failure

to respond of 1.19 (95%CI 0.77-1.84) with a 70% response in

the fresh FMT group and 75% in the frozen FMT group.

This increased to 85.6% and 90.7% respectively when

patients were given multiple infusions FMT due to lack of

response.

(B) Single infusion

Heterogeneity between groups: P=.368Overall (I^2 = 84.45%, P=.00);

Hamilton 2012 [60]

Aas 2003 [33]

Subtotal (I^2 = 76.41%, P=.00)

Yoon 2010 [41]

Lee 2014 [63]

Kao 2016 [26]Cammarota 2015 (FMT arm) [23]

Satokari 2015 [40]

Kelly 2012 [36]

RCT

Van Nood 2013 (FMT arm of RCT) [22]Youngster 2014 (Both FMT arms) [71]

Dutta 2014 [43]

Rohlke 2010 [38]

Kelly 2014 [30]

Brandt 2012 [68]Allegretti 2014 [42]

Patel 2013 [46]

Kelly 2016 (donor FMT arm) [27]Lee 2016 (Both FMT arms of RCT) [24]

Rubin 2013 [39]

Vigvari 2014 [72]

Agrawal 2016 [44]

MacConnachie 2009 [64]

Ray 2014 [37]

Kronman 2015 [45]Khan 2014 [62]

Case Series

Pathak 2014 [65]

Kassam 2012 [61]

Author

Zainah 2015 [67]

Tauxe 2016 [66]

Garborg 2010 [35]

Costello 2015 [69]

Ganc 2015 [34]

Subtotal (I^2 = 90.59%, P=.00)

Mattila 2012 [47]

Emmanuelson 2014 [70]

0.84 (0.79, 0.89)

0.86 (0.72, 0.95)

0.94 (0.70, 1.00)

0.86 (0.80, 0.90)

1.00 (0.74, 1.00)

0.48 (0.37, 0.58)

0.95 (0.84, 0.99)0.65 (0.41, 0.85)

0.96 (0.86, 1.00)

0.92 (0.75, 0.99)

0.81 (0.54, 0.96)0.70 (0.46, 0.88)

1.00 (0.87, 1.00)

0.95 (0.75, 1.00)

0.77 (0.67, 0.86)

0.88 (0.79, 0.95)0.86 (0.65, 0.97)

0.87 (0.69, 0.96)

0.91 (0.71, 0.99)0.52 (0.45, 0.58)

0.79 (0.68, 0.87)

0.90 (0.73, 0.98)

0.83 (0.76, 0.89)

0.73 (0.45, 0.92)

1.00 (0.83, 1.00)

0.90 (0.55, 1.00)0.90 (0.68, 0.99)

0.92 (0.62, 1.00)

0.81 (0.62, 0.94)

ES (95% CI)

0.57 (0.29, 0.82)

0.77 (0.59, 0.90)

0.73 (0.56, 0.85)

0.85 (0.62, 0.97)

0.83 (0.52, 0.98)

0.77 (0.56, 0.93)

0.90 (0.80, 0.96)

0.65 (0.43, 0.84)

100.00

3.22

2.60

82.17

2.36

3.49

3.222.77

3.28

2.94

2.602.77

2.97

2.77

3.45

3.442.83

3.03

2.833.65

3.43

3.03

3.59

2.55

2.77

2.212.77

2.36

2.97

Weight

2.49

3.05

3.18

2.77

2.36

17.83

3.41

2.86

%

0 .2 .4 .6 .8 1

Proportion responding

F IGURE 2 (Continued)

QURAISHI ET AL. | 485

Systematicreviewwithmeta-analysis:theefficacyofFMTforthetreatmentofrecurrentandrefractoryCDI

Quraishi et al, Aliment Pharmacol Ther 2017

of patients with IBD, solid organ transplants on immunosuppression,

HIV and cancers. The CDI cure rate observed after a single FMT

was 78%, with an overall cure rate of 89% following a second trans-

plant.

3.3.3 | Comparison between upper GI and lower GIroutes of delivery

The RCT that compared nasogastric (upper GI) vs colonic delivery of

FMT reported a cure rate of 60% at 8 weeks with a single infusion

and an overall cure rate 80% after second infusion when delivered

via NG.28 The cure rate with colonic delivery was 80% with a single

infusion at 8 weeks and an overall cure rate of 100% after a second

infusion.

Of the remaining studies, 25 case series and seven RCTs had

separate outcome data for modes of FMT delivery. Twenty-two

delivered FMT by the lower GI route (colonoscopy or retention

enema) and 11 delivered FMT by the upper gastrointestinal route

(upper GI endoscopy, nasogastric tube or naso-jejunal tube).

Results are displayed in Figure 3. The pooled response of CDI to

(A) Multiple infusions

Heterogeneity between groups: P=.790Overall (I^2=58.70%, P=.00);

Zainah 2015 [67]

Tauxe 2016 [66]

Ray 2014 [37]

Rubin 2013 [39]

Pathak 2014 [65]

RCT

Youngster 2014 (Both FMT arms) [71]

Cammarota 2015 (FMT arm) [23]

Rohlke 2010 [38]

Satokari 2015 [40]

Brandt 2012 [68]

Kao 2016 [26]

Fischer 2016 [59]

Kassam 2012 [61]

Yoon 2010 [41]

MacConnachie 2009 [64]

Van Nood 2013 (FMT arm of RCT) [22]

Author

Lee 2016 (Both FMT arms of RCT) [24]

Emmanuelson 2014 [70]Dutta 2014 [43]Costello 2015 [69]

Subtotal (I^2=.00%, P=.83)

Aas 2003 [33]

Youngster 2014 [28]

Patel 2013 [46]

Hamilton 2012 [60]

Allegretti 2016 [32]

Kelly 2014 [30]

Kelly 2016 (donor FMT arm) [27]

Subtotal (I^2=64.82%, P=.00)

Garborg 2010 [35]

Lee 2014 [63]

Agrawal 2016 [44]

Khan 2014 [62]

Vigvari 2014 [72]

Allegretti 2014 [42]

Case Series

Kelly 2012 [36]

Kronman 2015 [45]

Ganc 2015 [34]

Mattila 2012 [47]

0.92 (0.89, 0.94)

0.79 (0.49, 0.95)

0.87 (0.70, 0.96)

1.00 (0.83, 1.00)

0.79 (0.68, 0.87)

1.00 (0.74, 1.00)

0.90 (0.68, 0.99)

0.90 (0.68, 0.99)

1.00 (0.83, 1.00)

0.96 (0.86, 1.00)

0.91 (0.82, 0.96)

0.95 (0.84, 0.99)

0.81 (0.77, 0.85)

0.93 (0.76, 0.99)

1.00 (0.74, 1.00)

0.80 (0.52, 0.96)

0.94 (0.70, 1.00)

ES (95% CI)

0.88 (0.83, 0.92)

0.70 (0.47, 0.87)1.00 (0.87, 1.00)1.00 (0.83, 1.00)

0.91 (0.88, 0.94)

0.94 (0.70, 1.00)

0.90 (0.68, 0.99)

0.97 (0.83, 1.00)

0.95 (0.84, 0.99)

0.95 (0.74, 1.00)

0.85 (0.76, 0.92)

0.95 (0.77, 1.00)

0.92 (0.89, 0.95)

0.82 (0.67, 0.93)

0.86 (0.78, 0.92)

0.83 (0.76, 0.89)

1.00 (0.83, 1.00)

0.97 (0.83, 1.00)

0.86 (0.65, 0.97)

0.92 (0.75, 0.99)

1.00 (0.69, 1.00)

0.83 (0.52, 0.98)

0.94 (0.86, 0.98)

100.00

1.69

2.73

2.13

3.91

1.52

2.13

2.13

2.13

3.36

3.94

3.18

5.29

2.53

1.52

1.77

1.85

Weight

4.92

2.322.532.13

18.53

1.85

2.13

2.68

3.18

2.06

4.02

2.26

81.47

3.08

4.17

4.61

2.13

2.68

2.26

2.48

1.34

1.52

3.83

%

0 .2 .4 .6 .8 1

Proportion responding

F IGURE 2 Forest plot of the proportion responding to treatment for all included studies. A, Multiple infusions. ES (95% CI) is theproportion responding with its 95% confidence interval. B, Single infusion. ES (95% CI) is the proportion responding with its 95% confidenceinterval

484 | QURAISHI ET AL.Single infusion Multiple infusions

SafetyandefficacyoffecalmicrobiotatransplantationforrecurrentC.difficileinfectionfromaninternationalpublicstoolbank:Resultsfroma2,050patientmulti-centercohort

IDWeek2016

https://idsa.confex.com/idsa/2016/webprogram/Paper59497.html

•  482 healthcare facilities across 50 US states and 7 countries

•  Clinical cure rate across all delivery modalities and CDI patient populations was 84.0%

•  FMT by colonoscopy (85.8% clinical cure, n=1441) was superior to upper endoscopy (74.1% clinical cure, n=201) (p

AmJGastroenterol.2013Apr;108(4):478-98;

Treatment of ≥3 CDI recurrences If there is a third recurrence after a pulsed vancomycin regimen, FMT should be considered (Conditional recommendation, moderate quality evidence)

“For multiple recurrent CDI unresponsive to repeated antibiotic treatment, faecal transplantation in combination with oral antibiotic treatment is strongly recommended (A-I).” ClinMicrobiolInfect2014;20(Suppl.2):1–26

“This procedure should only be considered for patients with recurrent C. difficile infections that have failed to respond to antibiotics and other treatments.”

“Consider donor stool transplant in cases of recurrent CDI”

Survey of practice of FMT for CDI in the UK, 2015

N=255:120 Microbiology / ID84 Gastroenterology

Respondingsites England Scotland Wales

Totalsitesresponding 112 9 9

SitesthatperformedFMT 32 1 3

FMT>1year 17 1 3

FMT10patients 5 0 2

FMT

Summary

n Successrateof~90%whenFMTisusedtotreatrecurrentCDIn FMTisanappropriatesalvagetherapyformultiplyrecurrentCDIn Currently9RCTsreportedintheliteratureplusmanycaseserieswith1000’sofpatients

n  LowerGIadministrationappearstobemoreeffective,howeverismoreinvasiveandcostly

n FrozenFMTisaseffectiveasfresh,capsulisedproductalsoappearstobeeffectiveandmaybecomestandardrouteofadministration

n Significantdifferencesinmethodsandprotocolswithlackofstandardization–JointHIS/BSGguidelinesabouttogooutforconsultation

faecal microbiota transplants: the evidence and...

Documents