Endoscopic Duodenal MucosalResurfacing for the Treatment ofType 2 Diabetes: 6-Month InterimAnalysis From the First-in-HumanProof-of-Concept StudyDiabetes Care 2016;39:2254–2261 | DOI: 10.2337/dc16-0383

OBJECTIVE

To assess procedural safety and glycemic indices at 6 months in a first-in-humanstudy of duodenal mucosal resurfacing (DMR), a novel, minimally invasive, upperendoscopic procedure involving hydrothermal ablation of the duodenal mucosa,in patients with type 2 diabetes and HbA1c ‡7.5% (58 mmol/mol) on one or moreoral antidiabetic agents.

RESEARCH DESIGN AND METHODS

Using novel balloon catheters, DMRwas conducted on varying lengths of duodenumin anesthetized patients at a single medical center.

RESULTS

A total of 39 patients with type 2 diabetes (screening HbA1c 9.5% [80 mmol/mol];BMI 31 kg/m2) were treated and included in the interim efficacy analysis: 28 had along duodenal segment ablated (LS;∼9.3 cm treated) and 11 had a short segmentablated (SS; ∼3.4 cm treated). Overall, DMR was well tolerated with minimalgastrointestinal symptoms postprocedure. Three patients experienced duodenalstenosis treated successfully by balloon dilation. HbA1c was reduced by 1.2% at6 months in the full cohort (P < 0.001). More potent glycemic effects were ob-served among the LS cohort, who experienced a 2.5% reduction in mean HbA1c at3months postprocedure vs. 1.2% in the SS group (P< 0.05) and a 1.4% reduction at6 months vs. 0.7% in the SS group (P = 0.3). This occurred despite net medicationreductions in the LS cohort between 0 and 6 months. Among LS patients with ascreening HbA1c of 7.5–10% (58–86 mmol/mol) and on stable antidiabetic medi-cations postprocedure, HbA1c was reduced by 1.8% at 6 months (P < 0.01).

CONCLUSIONS

Single-procedure DMR elicits a clinically significant improvement in hyperglyce-mia in patients with type 2 diabetes in the short-term, with acceptable safety andtolerability. Long-term safety, efficacy, and durability and possible mechanisms ofaction require further investigation.

1Fractyl Laboratories, Inc., Waltham, MA2Vanderbilt University Medical Center, Nashville,TN3Brigham and Women’s Hospital, Boston, MA4Massachusetts General Hospital, Boston, MA5King’s College London, London, U.K.6Catholic University of Rome, Rome, Italy7CCO Clinical Center for Diabetes, Obesity andReflux, Santiago, Chile8Gastro Obeso Center, São Paulo, Brazil9Florida International University, Miami, FL

Correspondingauthor: Harith Rajagopalan, harith@fractyl.com.

Received 23 February 2016 and accepted 8 June2016.

Clinical trial reg. no. NCT01927562, clinicaltrials.gov.

© 2016 by the American Diabetes Association.Readers may use this article as long as the workis properly cited, the use is educational and notfor profit, and the work is not altered. More infor-mation is available at http://www.diabetesjournals.org/content/license.

See accompanying article, p. 2108.

Harith Rajagopalan,1

Alan D. Cherrington,2

Christopher C. Thompson,3 LeeM. Kaplan,4

Francesco Rubino,5 Geltrude Mingrone,6

Pablo Becerra,7 Patricia Rodriguez,7

Paulina Vignolo,7 Jay Caplan,1

Leonardo Rodriguez,7 and

Manoel P. Galvao Neto8,9

2254 Diabetes Care Volume 39, December 2016

EMER

GINGTECHNOLO

GIESANDTH

ERAPEU

TICS

Type 2 diabetes has reached epidemicproportions in Westernized countries.The current standard of care includeslifestyle or behavioral modification first,frequently coupled with the subsequentuse of an array of oral and injectablemedications. Despite a wide choice ofpharmacological interventions, manypatients do not achieve adequate con-trol of hyperglycemia (1). Poor compli-ance with complex medical regimens (2)and the fact thatmost available pharma-ceutical approaches do not adequatelyaddress underlying pathophysiologicaldefects may explain the limited efficacyof current therapies.Certain forms of bariatric surgery, es-

pecially those that involve bypass of theupper intestine, can exert powerful cor-rective effects on metabolism in obesesubjects with type 2 diabetes (3–6). Re-cently, there has been a groundswell ofinterest in establishing this form of inter-vention as an additional approach tomanaging type 2 diabetes beyond life-style modification or pharmacological ap-proaches. A new consensus statementcrafted by an international group of dia-betes experts and embraced by multipleprofessional organizations advocates thatsuch metabolic surgeries should be in-cluded among current treatment guide-lines for patients with type 2 diabetes(7). Experimental evidence demonstratesthat intestinal bypass surgery has directeffects on glucose metabolism, highlight-ing the importance of the small intestine,particularly the duodenum, in the physi-ology and pathophysiology of glucosehomeostasis (8,9). Anatomically, theduodenum is the first site of fuel recog-nition at the time of nutrient intake. Ob-servations in animal models and humansreveal that the duodenal mucosa ex-hibits abnormal hypertrophy and endo-crine hyperplasia in the presence ofdiabetes (10). Moreover, gastrointesti-nal bypass procedures or device-basedinterventions (i.e., the endoluminalsleeve) that prevent contact betweenduodenal mucosa, bile, and nutrients im-prove insulin sensitivity (11) and b-cellfunction (12). Further, in a rat modeland in patients with type 2 diabeteswho have undergone Roux-en-Y gastricbypass surgery, the improvement of glu-cose tolerance is quickly reversed whenthe bypassed duodenum is acutelyre-exposed to nutrients introduced viagastrostomy in the remnant stomach

(13,14). These observations underscorethe critical glucoregulatory role of the up-per intestine. More specifically, exclusionof nutrient contact from an abnormal du-odenal surface in type2diabetesmayelicitbeneficial downstream effects on metab-olism, perhaps through the reduction ofputative anti-incretin mechanisms (15).

Duodenal mucosal resurfacing (DMR)is a novel, minimally invasive, catheter-based upper endoscopic procedure in-volving hydrothermal ablation of theduodenal mucosa and subsequent mu-cosal healing. DMR could recapitulate,less invasively, some of the mechanismsof action of gastrointestinal bypass sur-gery. It may offer a new treatment ap-proach for type 2 diabetes by alteringthe duodenal mucosal surface itself,thereby altering downstream signalingand eliciting metabolic improvement.Preclinical studies conducted in theGoto-Kakizaki rat, a rodent model equiv-alent of human type 2 diabetes, supportthis thesis by demonstrating that selec-tive denudation of the duodenal mucosaconducted by an abrasion device resultedin immediate lowering of glycemia dur-ing an oral glucose gavage when com-pared with preprocedure levels andalso to a sham-treated group. Moreover,similarly conducted studies in a nondia-betic rodent model (Sprague-Dawley)showed no lowering of glycemia, indi-cating that this perturbation of the du-odenal mucosa reduced abnormalhyperglycemia but did not impact nor-moglycemia (unpublished data, FractylLaboratories, Inc.). Additional studiesin a pig model also demonstrate thatthe DMR procedure achieves a predict-able ablation of the intestinal mucosasurface without damage to the under-lying muscularis mucosa or deeperstructures (unpublished data, FractylLaboratories, Inc.). We therefore con-ducted a first-in-human clinical study ofsingle-procedure DMR in patients withtype 2 diabetes. Herewe report on safety,tolerability, and effectiveness from the6-month interim analysis.

RESEARCH DESIGN AND METHODS

Study DesignThis was a phase I, first-in-human, open-label, proof-of-concept study with asingle-arm, nonrandomized design per-formed at a single center in Santiago,Chile (CCO Clinical Center for Diabetes,

Obesity and Reflux). All procedureswere performed by physicians trainedin endoscopy (L.R., P.B., and M.P.G.N.)between August 2013 and November2014. The study protocol defines a pro-jected follow-up of 3 years, and we re-port here data from the first 6 monthsof postprocedure follow-up.

Study OversightThe study protocol was approved by anindependent ethics committee (Metropol-itano Comite de Etica Cientifico Oriente)in Santiago, Chile, and complied withthe recommendations of the Declarationof Helsinki. All patients provided writteninformed consent.

Procedural Development and Lengthof Treated Duodenal SegmentDuring a period of procedural develop-ment from August 2013 to December2013 (dose escalation phase), consecu-tive patients were treated with a singleapplication of thermal energy ablationover a maximum of 3 cm of duodenumlength with complete circumferentialapplication. The length of the ablatedregion and number of applications dur-ing a single procedure were increasedin subsequent procedures over thecourse of the study, with the goal ofachieving complete circumferential abla-tion of the postpapillary duodenum, cov-ering ;12 cm in length. From January2014 to November 2014, at least 9 cmof the duodenum was treated in all pa-tients, except when precluded by duode-nal anatomy.

PatientsEligible patients were adults with type 2diabetes aged 28–75 years, with a BMI24–40 kg/m2 and HbA1c 7.5–12% (58–108 mmol/mol) on at least one oralantidiabetic agent. Additional eligibil-ity criteria included fasting c-peptide.1 ng/mL and type 2 diabetes diagnosismade within 10 years prior to enroll-ment. Patients were excluded if theyhad type 1 diabetes (including anti-GAD positivity), current use of injectableantidiabetic medication, history ofprevious gastrointestinal surgery oranatomical abnormalities that wouldpreclude the DMR procedure, treat-ment with antiplatelet agents that couldnot be temporarily discontinued, orpregnancy.

care.diabetesjournals.org Rajagopalan and Associates 2255

Study ProcedureDMR is an endoscopic treatment consist-ing of intestinal luminal sizing, submuco-sal expansion with saline (designed toprovide a uniform ablative surface and athermally protective layer of saline be-tween the ablated mucosa and deeper tis-sue layers), and circumferential thermalablation along a length of the duodenum.In the current study, novel polyethyleneterephthalate balloon treatment cathe-ters (Revita system; Fractyl Laboratories,Inc.) were introduced into the duodenumvia a trans-oral endoscopic approach inanesthetized patients. The first catheterwas used to determine the size of theduodenum and inject saline into thesubmucosal space via three vacuum-assisted needle injectors oriented at1208 from one another around the cir-cumference of the balloon. Circumferen-tial mucosal lift was performed along thelength of the postpapillary duodenumfrom 1 cm distal to the ampulla of Vater(hepatopancreatic ampulla) to proximalto the ligament of Treitz. After removalof the initial catheter, a second ballooncatheter was introduced to perform ther-mal ablation on the lifted area. Under di-rect endoscopic visualization, discretecircumferential thermal ablations of;10 s eachwere applied at temperaturesof ;908C to obtain up to five longitudi-nally separated ablations along the lengthof the postpapillary duodenum. Care wastaken to avoid the ampulla of Vater toprevent damage to the biliary tree andto avoid treatment in or beyond the liga-ment of Treitz.Patients were discharged within 24 h

after the procedure and prescribed a pro-gressive diet (liquids → pureed foods →soft foods) for 2 weeks. Although nospecific recommendations were made,physicians were requested to minimizealterations in antidiabetic medicationsexcept when medically indicated. Med-ications and doses were recorded bythe investigators at all follow-up timepoints. Background antidiabetic medi-cation usewas assigned into “stable,” “in-creased,” and “decreased” categoriesover time to allow subset analysis ofthe patient cohort.

Outcome Measures and AssessmentsThe first-in-human study evaluated pro-cedural safety and efficacy. Efficacy wasassessed using mixed-meal tolerancetesting (MMTT) and measuring fasting

plasma glucose (FPG), fasting plasma in-sulin, and HbA1c.

Comparisons were made betweenpreprocedure (screening) and postpro-cedure metabolic parameters and alsobetween patient cohorts based on thelength of the treated duodenal seg-ment (short vs. long), HbA1c levels, andchanges in background medication use.

Preproceduremeasurements at screen-ing consisted of patient history, vitalsigns, physical examination, medicationreview, blood analysis (including HbA1c,fasting glucose, fasting insulin, choles-terol, and other pancreatic and livermarkers), and MMTT. All participantsunderwent a screening endoscopy, with afollow-up endoscopy 3 months after thetreatment procedure.

Patients were seen on days 7 and14 after the procedure for a physicalexam, standard blood analysis, collec-tion of data on medication use, and sur-veillance for adverse events. Theseassessments and MMTT were also per-formed at 1, 3, and 6 months after theprocedure, with the exception of repeatendoscopy and duodenal biopsy, whichwas performed within 3 months to as-sess mucosal healing in the first cohortof treated patients.

Statistical AnalysisAll statistical tests were two sided at thea level of 0.05 unless stated otherwise.No adjustments for multiple hypothesistesting were made. For the efficacy

analyses, a mixed model with repeatedmeasures was used to analyze changefrom preprocedure levels when morethan two postprocedure measures wereassessed. Safety data were tabulated bypatient and events overall and by severityand relationship to device, including thenumber of overall adverse events, seriousadverse events, and unanticipated ad-verse device effects by severity. Whereapplicable, comparisons are reported asmeans 6 SEM. All statistical analyseswere performed using Statistical AnalysisSystem (SAS) release 9.4 or higher (SAS,Cary, NC).

RESULTS

We report data for the first 44 patientsenrolled in the study who underwentscreening endoscopy in preparation forthe DMR procedure (intent-to-treatpopulation). Among these patients, fourdid not receive thermal ablation (twofailed screening endoscopy, one had tor-tuous anatomy, and one had the proce-dure stopped prior to ablation to preventprolonged anesthesia), and one subjectunderwent the procedure but was sub-sequently excluded from the efficacyanalysis (although included in the safetyanalysis) due to anti-GAD positivity in-dicative of type 1 diabetes documentedafter the index procedure. One addi-tional treated patient withdrew consentbefore the final 6-month visit but wasincluded in both the safety and efficacyanalyses. Clinical characteristics of the

Table 1—Clinical characteristics at screening (intent-to-treat population)

Patient characteristics Value (n = 44)

Age, years (range) 53.4 6 7.5 (38–65)

Sex, n (%)Female 16 (36)Male 28 (64)

Weight (kg) 84.4 6 11.9

Height (cm) 165.3 6 8.4

BMI (kg/m2) 30.8 6 3.5

Systolic blood pressure (mmHg) 122.0 6 14.2

Diastolic blood pressure (mmHg) 77.0 6 8.1

Duration of type 2 diabetes, years (range) 5.7 6 2.2 (0.2–9.7)

HbA1c% 9.6 6 1.4mmol/mol 81 6 16

FPG (mg/dL) 187 6 58

Oral antidiabetic medicationsMetformin 42 (98)Sulfonylurea 16 (37)

Data are mean 6 SD or n (%), unless otherwise indicated.

2256 Duodenal Mucosal Resurfacing in Type 2 Diabetes Diabetes Care Volume 39, December 2016

intent-to-treat population at screeningare presented in Table 1.A total of 39 treated patients were in-

cluded in theefficacy analysis. The averagetime between screening and treatment inthis efficacy cohort was 5.5 weeks.All treated patients received circumfer-

ential ablation of between 3 and 15 cm ofthe postpapillary duodenal mucosa. Themean procedure time from beginning ofsubmucosal saline expansion to comple-tion of thermal ablation was 54 min(interquartile range 47–69). The meanlength of treated duodenal mucosa inthe full cohort was 7.8 cm. Over thecourse of the study, 28 patients hadablation of a long segment of duodenalmucosa (LS-DMR; $9 cm of ablation)and 11 had a short segment ablated(SS-DMR; ,6 cm of ablation). SS-DMRwas performed either in the course ofearly procedural development or laterdue to complicated duodenal anatomy.The average length of duodenal mucosaablated was 9.3 cm in LS-DMR and3.4 cm in SS-DMR.

Safety and TolerabilityThe DMR procedure was completedwithout complication in all 40 treatedpatients. There was no gastrointestinalbleeding, perforation, pancreatitis, se-vere hypoglycemia, or evidence of mal-absorption (i.e., calcium abnormalitiesor iron deficiency anemia), either in theperiod immediately following the proce-dure or at later follow-up visits. No pa-tients experienced difficulty tolerating

the oral diet in the days immediately fol-lowing the procedure.

The most common study-related ad-verse event was transient, postproce-dural abdominal pain (in 8 of 40 patients)due to air insufflation and/or endotrachealintubation, for which analgesic medica-tions were neither administered nor re-quired. No patient reported experiencingdiscomfort by 48 h after the procedure.

Three patients developed a duodenalstenosis that presented as epigastric painand vomiting 2–6 weeks after the proce-dure. These patients were treated withendoscopic balloon dilation with full res-olution of symptoms and no furthersequelae. Root cause analysis revealedassignable faults in each case for whichdevice improvements and proceduralchanges have since been instituted.

Follow-up endoscopies and duodenalbiopsies (1 month: endoscopy [n = 19]and biopsy [n = 19]; 3 months: endos-copy [n = 39] and biopsy [n = 9]) showedmucosal healing in all evaluated pa-tients. Figure 1 shows the appearanceof the duodenal mucosa prior to theprocedure, immediately after hydro-thermal ablation, and 1 month afterthe procedure as seen during follow-upendoscopy. No patients experiencedany signs of infection.

EfficacyDMR elicited significant improvementsin glycemic indices. FPG reductionswere noted within 1 week of the pro-cedure (Fig. 2A) and HbA1c reductionswere observed as early as 1 month

(Fig. 3) after DMR and were still pre-sent in a significant proportion of sub-jects (29 of 39 patients) at 6 months offollow-up. These reductions were ob-servedwithout a change in fasting plasmainsulin (screening 11.7 6 1.0 mIU/L,3months 11.86 1.5mIU/L, and 6months11.6 6 1.3 mIU/L for LS-DMR cohort[n = 28]).

HbA1c was reduced by 1.2 6 0.3% at6 months in the full cohort (P , 0.001).More potent glycemic effects were ob-served in LS-DMR patients (Fig. 2A).LS-DMR patients experienced a 2.5 60.2%reduction inmeanHbA1cat3months,compared with a 1.26 0.5% reductionamong the SS-DMR cohort at the sametime point (P, 0.05 between groups).This trend was continued through6 months of observation, with a meanHbA1c reduction of 1.4 6 0.3% in theLS-DMR cohort and 0.7 6 0.5% in theSS-DMR cohort (P = 0.3 between groups).This occurred despite net medication re-ductions in the LS-DMR cohort between0 and 6 months.

Individual patient data plots for FPG forthe LS cohort show that the majority ofpatients had a robust glucose-loweringresponse after DMR (Fig. 2B). MMTT re-sults suggest that the improvement inglycemic indices manifested through im-provements in both fasting and post-prandial glycemia (Fig. 2C and D), withthe majority of the effect seemingly at-tributable to improvements in fastinghyperglycemia. The procedure also ap-peared to exert glycemic improve-ments over a wide range of screening

Figure 1—The duodenal mucosa prior to DMR (A), immediately after hydrothermal ablation (B), and 1 month after the procedure (C) as seen duringfollow-up endoscopy.

care.diabetesjournals.org Rajagopalan and Associates 2257

HbA1c values, including values of 10%(86 mmol/mol) and greater (Fig. 3A). How-ever, some erosion of the glycemic effectwas observed between the 3- and 6-monthfollow-up, as seen in Fig. 2C and Fig. 3.During the 6-month follow-up period,

concomitant medication use changedin some patients in both the SS (3 in-creased, 3 stable, and 5 decreased)and LS cohorts (0 increased, 14 stable,and 14 decreased), with a trend towardgreater improvements in glycemic con-trol in patients whose usage of antidia-betic medication remained stable (Fig.3B and Fig. 4) (P = 0.11 at 6 months).Among LS patients with a screeningHbA1c 7.5–10% (58–86 mmol/mol) andon stable antidiabetic medications post-procedure (n = 8), HbA1c was reduced by1.8 6 0.5% at 6 months (P , 0.01) and

was accompanied by a modest weightreduction of 3.9 6 0.5 kg at 3 months(P, 0.001) and 2.56 0.1 kg at 6months(P , 0.05). There was no statisticallysignificant correlation between degreeof weight loss and magnitude of HbA1cimprovement.

CONCLUSIONS

In this first-in-human study, a singleprocedure endoscopic DMR ablationelicited a substantial improvement inglycemia in medically treated patientswith suboptimally controlled type 2 di-abetes followed for 6 months, with anacceptable safety and tolerability pro-file observed to date.

DMR appeared to exhibit dose depen-dency, with LS ablation exerting more

potent glycemic effects, as seen by the

statistically significant difference in gly-cemic effect between SS-DMR versusLS-DMR at 3 months (P , 0.05). This isstated with some caution, as the studywas not designed to formally examineablation dose dependency (i.e., SS vs.LS ablation), and the optimal length ofablated duodenum requires additionalstudy. As one would anticipate with anovel, procedure-based intervention,this first-in-human study was conductedin an iterativemanner where proceduralfeasibility and patient safety were ofprime importance in the earlier cases.As procedural expertise increased, LSablation became more feasible, butwith ongoing attention to patientsafety. In the LS cohort, DMR exertedimprovements in glycemia as early as1–2 weeks postprocedure, similar tothe prompt effects seen after bariatricsurgery and suggesting the possibility ofsimilar mechanisms of action (9,16). Pa-tients in this study were asked to adhereto only modest dietary modification inthe first 2 weeks after DMR, and no rec-ommendations on diet or caloric restric-tion were made beyond that time.Therefore, the early improvement in gly-cemic control induced by DMR is un-likely to be explained by decreasedcaloric intake. In addition, the lack of sub-stantial weight loss through 6 months offollow-up also suggests that decreasedcaloric intake is not a major mechanism.

Beyond the early improvement in gly-cemia, DMR led to plasma glucose andHbA1c improvement over the 6-monthfollow-up. Meal challenge data suggestthat the predominant effect of DMR wason fasting hyperglycemia, with a lessstriking improvement in postprandialexcursion. It is interesting that a gut-related intervention targeted at theduodenum would exert predominant ef-fects on fasting, not postprandial, glyce-mia. This suggests a potential effect onhepatic glucose production, possiblythrough an insulin-sensitizing mechanism,in line with previous observations induodenal-jejunal bypass (17) and Roux-en-Ygastric bypass surgery (18–20). The overallglucose-lowering effect of DMR appearsless potent than that observed with bari-atric surgery, but the intended targetedintervention of DMR (an alteration of nu-trient exposure to the duodenal surface)would only approximate one of the mul-tiple anatomical modifications that comewith Roux-en-Y gastric bypass surgery. It is

Figure 2—Effect of DMR on ambient glycemia. A: Effect of SS (white circles) and LS (black circles)DMR treatment on FPG plotted to 3 months (n = 39). B: FPG change from screening plotted to3 months in individual subjects who received LS-DMR (n = 28). C: Meal challenge plasma glucose(PG) from fasting to 120 min after meal ingestion in LS-DMR subjects at screening (whitesquares, n = 28), 3 months (black squares, n = 27) (no 3-month MMTT data for one subject),and 6 months (black circles, n = 28). D: Change from fasting in area under the curve (AUC) forMMTT at screening, 3 months, and 6 months in LS-DMR subjects (n = 28). Values for A, C, and Dare reported as mean 6 SEM. mo, month; S, preprocedure levels (screening).

2258 Duodenal Mucosal Resurfacing in Type 2 Diabetes Diabetes Care Volume 39, December 2016

interesting nevertheless that a targetedintervention by endoscopic procedurecan elicit glycemic improvement withoutthe more extensive anatomical disruptionof surgical approaches. Future develop-ment of DMR will focus on the underlyingmechanism and efforts to more predict-ably manifest a metabolic benefit.Examination of the overall glycemic ef-

fects in this study suggests some erosionof effect in the latter phase of study. Acloser examination of glycemic responseby individual patient in the LS cohortwith pretreatment HbA1c levels of#10%(86 mmol/mol) (Fig. 4) shows thatwhereas some patients experience an

almost complete erosion of the effectby the 6-month follow-up (6 of 18 LS sub-jects with pretreatment HbA1c #10%[86 mmol/mol] had ,0.5% reduction inHbA1c at 6 months), many patientsmanifest a sustained effect on glyce-mia throughout the 6-month period(10 of 18 LS subjects with pretreat-ment HbA1c #10% [86 mmol/mol] hadan HbA1c ,7.5% [58 mmol/mol] at6 months). In addition, erosion of theglycemic effect in some cases canbe explained by a withdrawal of or non-adherence to prescribed concomitantantidiabetic medication. This includesthe single patient in Fig. 4 who had

reduced antidiabetic medication useand experienced a significant deteriora-tion in glycemic control.

Themore precisemechanism throughwhich DMR elicits beneficial metaboliceffects likely relates to the pathogenicchanges observed in the gut of sub-jects with type 2 diabetes. The duodenalmucosa becomes abnormal with highfat/hexose feeding (10,21–25), whichin turn disturbs local nutrient absorp-tion (26) and neuronal and hormonal sig-naling (24,25,27). This appears to alsoinvolve the overgrowth of certain mu-cosal cell types, including the K cell(10,23), which is recognized for its glu-coregulatory role in secreting the incretinhormoneGIP. Duodenal biopsy from hu-mans with diabetes also shows hyper-trophy, thickening, and an overgrowthof entero-endocrine cell types (22). In ad-dition, an as yet unidentified factor iso-lated from proximal intestinal cells maydirectly impact systemic insulin sensi-tivity, as myocytes become insulin resis-tant when exposed in vitro to proteinsproduced by duodenal/jejunal mucosafrom subjects with diabetes (28). More-over, studies in animals and humansshow that duodenal exclusion ame-liorates the metabolic disturbance(9,29,30) of type 2 diabetes, and in hu-mans and rats, an intestinal device thatprevents nutrient contact with the du-odenal mucosa also leads to improve-ment in metabolic measures (31–33).

The Revita DMR procedure appearedto be safe and well tolerated in thisfirst-in-human trial. The procedure con-sists of duodenal sizing, saline expansionof the submucosal space, and hydrother-mal treatment of the mucosa at denatur-ation temperatures to ablate superficiallayers and trigger a rejuvenative healingresponse. The DMR procedure is analo-gous to the commonly performed radio-frequency ablation of the esophagus,where ablation and subsequent healingof esophageal mucosa is used to treatBarrett esophagus and esophageal dys-plasia (34,35). DMR treatment targetsthe mucosal surface of the duodenumdistal to the ampulla of Vater and proxi-mal to the ligament of Treitz. There weregastrointestinal symptoms noted post-procedure, but these were transient andof mild or moderate severity. There wereno cases of perforation, pancreatitis, orbleeding and no apparent evidence ofmalabsorption. Three cases of duodenal

Figure 3—DMR effect on HbA1c.A: Mean change in HbA1c in LS-DMR subjects with higher (.10%[86 mmol/mol], white squares, n = 10) and lower (#10% [86 mmol/mol], black squares, n = 18)pretreatment HbA1c levels. B: Effect of LS-DMR in subjects with lower (#10% [86 mmol/mol],n = 18) pretreatment HbA1c levels where background antidiabetic medication remained stable(black circles, n = 8) or was reduced (white circles, n = 10) during the 6-month follow-up period(for one subject, 6-month medication data were not recorded). All values are reported asmean 6 SEM. S, preprocedure levels (screening).

Figure 4—HbA1c in individual subjects who received LS-DMR and had lower (#10% [86mmol/mol])pretreatment HbA1c levels plotted to 6 months (n = 18) and displayed as absolute HbA1c levels(hatched line indicates American Diabetes Association treatment goal of 7%) (A) and change frompretreatment HbA1c levels (B). Individual subject antidiabetic medication status is represented asstable (black circles) or reduced (white circles).

care.diabetesjournals.org Rajagopalan and Associates 2259

stenosis were reported, occurring within6 weeks of the procedure and presentedwith epigastric pain and vomiting. Allwere addressed with endoscopic balloondilatation with no longer-term sequelae.Since that time, the catheter system hasbeen further optimized to reduce the riskof stenosis, but this is a particular safetyevent that will require further surveil-lance. On an additional safety note,DMR does not appear to manifest poten-tial for inducing hypoglycemia within thefirst 6 months after the procedure, as thiscomplication was not observed in the ab-sence of background hypoglycemic med-ication use, despite an overall reductionof ambient glycemia.Any first-in-human study comes with

shortcomings, in part because aspects ofthe intervention are new and untested.That stated, the outcome of this first-in-human study is cautiously optimistic.This unique, single-procedure interventionnot only exerts robust effects on glyce-mia in patients with type 2 diabetes butdoes sowithminimal perturbation to thepatient and with a relatively reassuringsafety profile to date. The majority ofpatients were able to receive the full in-tended ablation, but full ablation withDMR was not feasible in some patientsowing to anatomical limitations.Our findings suggest that minimally in-

vasive upper gastrointestinal interventionthrough DMR can improve glycemia intype 2 diabetes and represents an inter-esting potential adjuvant or alternative topharmacological treatment. The DMR ap-proach may overcome treatment adher-ence and compliance issues, a majorshortcoming of all pharmacological ap-proaches. DMR also provides a windowinto the intriguing and specific role ofthe duodenum in regulating downstreammetabolism. Further work is necessary tobetter understand the clinical utility ofthis procedure-based intervention in con-trolled trial conditions in larger numbersof patients.

Acknowledgments. The authors thank CraigGardner (Fractyl Laboratories, Inc.) for prepara-tion of figures and Katherine Kacena (KatherineKacena Consulting, Natick, MA) for statisticalanalysis.Funding. Funding for this study was providedby Fractyl Laboratories, Inc.Duality of Interest. H.R. and J.C. are employeesof Fractyl Laboratories, Inc. and own shares in thecompany. A.D.C. is a scientific advisor/consultant

to Biocon, Eli Lilly and Company, Fractyl Labora-tories, Inc., Hanmi, Islet Sciences, Merck, Meta-vention, Novo Nordisk, NuSirt Biopharma, ProfilInstitute for Clinical Research, Inc., Sensulin, SilverLake, ThermalinDiabetes, Thetis Pharmaceuticals,vTv Therapeutics, ViaCyte, Viking, and Zafgen;receives research support from Eli Lilly and Com-pany, Merck, Metavention, Novo Nordisk, SilverLake, and Thermalin Diabetes; and has equity/stock options in Fractyl Laboratories, Inc., Meta-vention, Sensulin, Thetis Pharmaceuticals, andZafgen. C.C.T. is a scientific advisor/consultantto Apollo Endosurgery, Covidien, Fractyl Labora-tories, Inc., Olympus, and USGI Medical; receivesresearch support fromApollo Endosurgery, AspireBariatrics, and USGI Medical; receives labora-tory support from Olympus; and has an owner-ship interest in GIWindows. L.M.K. is a scientificadvisor/consultant to Fractyl Laboratories, Inc.,GI Dynamics, Janssen, MedImmune, and NovoNordisk. F.R. is a scientific advisor/consultant toEthicon, Fractyl Laboratories, Inc., Medtronic,and NGM Biopharmaceuticals and has receivedspeaker honoraria from Sanofi. G.M. is a con-sultant for Novo Nordisk and Fractyl Laborato-ries, Inc. and has received research funds fromAstraZeneca, Fractyl Laboratories, Inc., andMetacure. P.B., P.R., P.V., and L.R. have receivedresearch support from Fractyl Laboratories, Inc.M.P.G.N. serves on the scientific advisory boardof and receives research funding from GI Dy-namics and Fractyl Laboratories, Inc. No otherpotential conflicts of interest relevant to thisarticle were reported.AuthorContributions.H.R. participated in thedesign of the study and the data analysis andwrote, edited, reviewed, and approved themanuscript. A.D.C. and L.M.K. provided criticalreview of the manuscript and reviewed, edited,and approved the manuscript. C.C.T. and F.R.provided critical review of the manuscript andedited and approved the manuscript. G.M.participated in the design of the study, providedcritical review, and edited and approved themanuscript. P.B. and L.R. performed the studyprocedures and reviewed, edited, and approvedthemanuscript. P.R. served as study coordinatorat the study site, assisted during study proce-dures, and reviewed, edited, and approved themanuscript. P.V. saw patients at the site forscreening and follow-ups, made determina-tions regarding medication management, andreviewed, edited, and approved themanuscript.J.C. participated in the design of the study,contributed to the manuscript draft, providedcritical review, and edited and approved themanuscript. M.P.G.N. performed the study pro-cedures, provided critical review, and edited andapprovedthemanuscript.H.R. is theguarantorofthis work and, as such, had full access to all thedata in the study and takes responsibility for theintegrity of the data and the accuracy of the dataanalysis.Prior Presentation. Parts of this study werepresented at the 3rd World Congress on Inter-ventional Therapies for Type 2 Diabetes/2ndDiabetes Surgery Summit DSS-II, London, U.K.,28–30 September 2015, and at the 13th AnnualWorld Congress on Insulin Resistance, Diabetesand Cardiovascular Disease, Los Angeles, CA,19–21 November 2015.

References1. Inzucchi SE, Bergenstal RM, Buse JB, et al.Management of hyperglycemia in type 2 diabe-tes, 2015: a patient-centered approach: updateto a position statement of the American Diabe-tes Association and the European Associationfor the Study of Diabetes. Diabetes Care 2015;38:140–1492. Egede LE, Gebregziabher M, Dismuke CE,et al. Medication nonadherence in diabetes:longitudinal effects on costs and potential costsavings from improvement. Diabetes Care 2012;35:2533–25393. Zervos EE, Agle SC, Warren AJ, et al. Amelio-ration of insulin requirement in patients under-going duodenal bypass for reasons other thanobesity implicates foregut factors in the patho-physiology of type II diabetes. J Am Coll Surg2010;210:564–572, 572–5744. Laferrere B, Reilly D, Arias S, et al. Differentialmetabolic impact of gastric bypass surgery ver-sus dietary intervention in obese diabetic sub-jects despite identical weight loss. Sci TranslMed 2011;3:80re25. Salinari S, le Roux CW, Bertuzzi A, RubinoF, Mingrone G. Duodenal-jejunal bypass andjejunectomy improve insulin sensitivity inGoto-Kakizaki diabetic rats without changesin incretins or insulin secretion. Diabetes2014;63:1069–10786. Jacobsen SH, Olesen SC, Dirksen C, et al.Changes in gastrointestinal hormone responses,insulin sensitivity, and beta-cell function within2weeks after gastric bypass in non-diabetic sub-jects. Obes Surg 2012;22:1084–10967. Rubino F, Nathan DM, Eckel RH, et al. Meta-bolic surgery in the treatment algorithm fortype 2 diabetes: a joint statement by interna-tional diabetes organizations. Diabetes Care2016;39:861–8778. Rubino F, Marescaux J. Effect of duodenal-jejunal exclusion in a non-obese animal modelof type 2 diabetes: a new perspective for an olddisease. Ann Surg 2004;239:1–119. Rubino F, Forgione A, Cummings DE, et al.Themechanism of diabetes control after gastro-intestinal bypass surgery reveals a role of theproximal small intestine in the pathophysiologyof type 2 diabetes. Ann Surg 2006;244:741–74910. Gniuli D, Calcagno A, Dalla Libera L, et al.High-fat feeding stimulates endocrine, glucose-dependent insulinotropic polypeptide (GIP)-expressing cell hyperplasia in the duodenumof Wistar rats. Diabetologia 2010;53:2233–224011. Ferrannini E, Mingrone G. Impact of differ-ent bariatric surgical procedures on insulin ac-tion and beta-cell function in type 2 diabetes.Diabetes Care 2009;32:514–52012. Klein S, Fabbrini E, Patterson BW, et al.Moderate effect of duodenal-jejunal bypass sur-gery on glucose homeostasis in patients withtype 2 diabetes. Obesity (Silver Spring) 2012;20:1266–127213. Dirksen C, Hansen DL, Madsbad S, et al.Postprandial diabetic glucose tolerance is nor-malized by gastric bypass feeding as opposed togastric feeding and is associated with exagger-ated GLP-1 secretion: a case report. DiabetesCare 2010;33:375–37714. Shimizu H, Eldar S, Heneghan HM, SchauerPR, Kirwan JP, Brethauer SA. The effect of selec-tive gut stimulation on glucose metabolism

2260 Duodenal Mucosal Resurfacing in Type 2 Diabetes Diabetes Care Volume 39, December 2016

after gastric bypass in the Zucker diabetic fattyrat model. Surg Obes Relat Dis 2014;10:29–3515. Rubino F, R’bibo SL, del Genio F, MazumdarM, McGraw TE. Metabolic surgery: the role ofthe gastrointestinal tract in diabetes mellitus.Nat Rev Endocrinol 2010;6:102–10916. Camastra S, Gastaldelli A, Mari A, et al. Earlyand longer term effects of gastric bypass surgeryon tissue-specific insulin sensitivity and beta cellfunction in morbidly obese patients with andwithout type 2 diabetes. Diabetologia 2011;54:2093–210217. Breen DM, Rasmussen BA, Kokorovic A,Wang R, Cheung GWC, Lam TKT. Jejunal nutrientsensing is required for duodenal-jejunal bypasssurgery to rapidly lower glucose concentrationsin uncontrolled diabetes. Nat Med 2012;18:950–95518. Umeda LM, Silva EA, Carneiro G, Arasaki CH,Geloneze B, Zanella MT. Early improvement inglycemic control after bariatric surgery and itsrelationships with insulin, GLP-1, and glucagonsecretion in type 2 diabetic patients. Obes Surg2011;21:896–90119. Jørgensen NB, Jacobsen SH, Dirksen C, et al.Acute and long-term effects of Roux-en-Y gas-tric bypass on glucose metabolism in subjectswith type 2 diabetes and normal glucose toler-ance. Am J Physiol Endocrinol Metab 2012;303:E122–E13120. Bojsen-Møller KN, Dirksen C, Jørgensen NB,et al. Early enhancements of hepatic and laterof peripheral insulin sensitivity combined withincreased postprandial insulin secretion con-tribute to improved glycemic control afterRoux-en-Y gastric bypass. Diabetes 2014;63:1725–1737

21. Adachi T, Mori C, Sakurai K, Shihara N,Tsuda K, Yasuda K. Morphological changes andincreased sucrase and isomaltase activity insmall intestines of insulin-deficient and type 2diabetic rats. Endocr J 2003;50:271–27922. Theodorakis MJ, Carlson O, Michopoulos S,et al. Human duodenal enteroendocrine cells:source of both incretin peptides, GLP-1 andGIP. Am J Physiol Endocrinol Metab 2006;290:E550–E55923. Bailey CJ, Flatt PR, Kwasowski P, Powell CJ,Marks V. Immunoreactive gastric inhibitorypolypeptide and K cell hyperplasia in obese hy-perglycaemic (ob/ob) mice fed high fat and highcarbohydrate cafeteria diets. Acta Endocrinol(Copenh) 1986;112:224–22924. Flatt PR, Bailey CJ, Kwasowski P, Swanston-Flatt SK. Effects of diets rich in sucrose, coconutfat and safflowerseed oil on the development ofthe obese hyperglycaemic (ob/ob) syndrome inmice. Diabetes Res 1990;13:23–2825. Ponter AA, Salter DN, Morgan LM, Flatt PR.The effect of energy source and feeding level onthe hormones of the entero-insular axis andplasma glucose in the growing pig. Br J Nutr1991;66:187–19726. Nguyen NQ, Debreceni TL, Bambrick JE,et al. Accelerated intestinal glucose absorptionin morbidly obese humans: relationship toglucose transporters, incretin hormones, andglycemia. J Clin Endocrinol Metab 2015;100:968–97627. Morgan LM, Hampton SM, Tredger JA,Cramb R, Marks V. Modifications of gastric in-hibitory polypeptide (GIP) secretion in man by ahigh-fat diet. Br J Nutr 1988;59:373–38028. Salinari S, Debard C, Bertuzzi A, et al. Jejunalproteins secreted by db/db mice or insulin-

resistant humans impair the insulin signalingand determine insulin resistance. PLoS One2013;8:e5625829. Irwin N, Flatt PR. Evidence for beneficialeffects of compromised gastric inhibitory poly-peptide action in obesity-related diabetes andpossible therapeutic implications. Diabetologia2009;52:1724–173130. Habegger KM, Al-Massadi O, Heppner KM,et al. Duodenal nutrient exclusion improvesmetabolic syndrome and stimulates villus hy-perplasia. Gut 2014;63:1238–124631. Mu~noz R, Carmody JS, Stylopoulos N, DavisP, Kaplan LM. Isolated duodenal exclusion in-creases energy expenditure and improves glu-cose homeostasis in diet-induced obese rats.Am J Physiol Regul Integr Comp Physiol 2012;303:R985–R99332. Kindel TL, Yoder SM, Seeley RJ, D’AlessioDA, Tso P. Duodenal-jejunal exclusion improvesglucose tolerance in the diabetic, Goto-Kakizakirat by a GLP-1 receptor-mediated mechanism.J Gastrointest Surg 2009;13:1762–177233. de Jonge C, Rensen SS, Verdam FJ, et al.Endoscopic duodenal-jejunal bypass liner rap-idly improves type 2 diabetes. Obes Surg 2013;23:1354–136034. Ganz RA, Utley DS, Stern RA, Jackson J, BattsKP, Termin P. Complete ablation of esophagealepithelium with a balloon-based bipolar elec-trode: a phased evaluation in the porcine andin the human esophagus. Gastrointest Endosc2004;60:1002–101035. Dunkin BJ, Martinez J, Bejarano PA, et al.Thin-layer ablation of human esophageal epi-thelium using a bipolar radiofrequency balloondevice. Surg Endosc 2006;20:125–130

care.diabetesjournals.org Rajagopalan and Associates 2261