Accepted Article Wei Li,MD China; China; People's Republic of China; 9 8 7 6 5 4 3 2 1 MD Jin Du,

Yiming Mu Cangzhou People’s Hospital, Changzhou, People's Republic of China. ofChina. Republic People's Changzhou, Hospital, People’s Cangzhou Ho General Force Police Armed People's Chinese Hospital,Jingzhou Jingzhou Central Wuhan 6 Hospita Affiliated University Medical Xuzhou China; Republic of People's Hebi, Co., Hospital, Ltd General (Group) Hebi Coal China; Hospital, People'sRepublicGongren Tangshan, of Tangshan China; of Republic Liaoning Hospital,People's Shenyang, People’s Chinese People's ofEndocrinology, Department Monotherapy: Results of a Phase IV ofaPhase Results Monotherapy: Efficacy andSafetyofSaxaglipti

article as doi: 10.1111 /dom.12942 differencesbetween version lead to this may been through the copyediting, typesetting, pagination andproofreading process, which This article hasbeen accepted for publication andundergone full peer review but has not Patients withType 2Diabetes th Hospital, Wuhan,People'sRepublicofChina; * 1 , MD, , MD, Ph.D *, Li Liang, MD *, LiLiang, 5 , Liya Shen, MD , Liya Shen, This article isprotected bycopyright. Allrights reserved. 1 2 Study (The SMARTStudy) *, Hui Fang, MD *, Hui Fang, 6 , Xueying Wang,MD , Xueying , People's Republic China; , People's of Mellitus Uncontrolled onMetforminMellitus Uncontrolled n Compared with Acarbose in Chinese inChinese Acarbose Comparedwith n

3 Open-Label Randomized Controlled Controlled Randomized Open-Label , Fengmei Xu,MD Liberation Army General Hospital, Beijing, Beijing, General Hospital, Army Liberation

l, Xuzhou, Jiangsu, of People'sRepublic l, Xuzhou, spital, Beijing, People's Republic of China; spital, People'sRepublic Beijing, and the Versionof Record. Please cite this 7 , Chun Xu, MD , ChunXu, 4 , 8 , FangBian,MD 9 , This articleis protectedbycopyright. Allrightsreserved. Accepted Article control (Week 24 HbA1c change: –0.82 and –0.78%, respectively; difference[95% respectively; and–0.78%, –0.82 change: control (Week24HbA1c baseline efficacy values recorded. Saxaglipti 243ofthesehad and and238 respectively, andacarbose, received patients 244 241and system); response voice/web (interactive system randomization Results: received who in all patients withoutGIAEs. achievingpatients HbA1c<7.0% with gastrointestinal ofpatients the proportion pr the 24included at Week assessed outcomes efficacy 24.Secondary atWeek HbA1c in frombaseline change was outcome primary Theat sitesin China conducted (24/09/2014–29/09/2015). 35 IVstudy open-label Phase Materials andMethods: with monotherapy. controlled inadequately (T2DM) mellitus 2 type with patients Chinese in acarbose Aim: Abstract * Email: [email protected] 7711 Fax: +86-10-6693 9001 Tel: +86-10-5549 China of People'sRepublic Beijing 100853, Army GeneralLiberation Hospital Chinese People's of Endocrinology, Department Yiming Mu * Corresponding author author Corresponding Jin Du and Li Liang had the same level of input and are co-firstare and level theauthors Li ofinput and Lianghad same Jin Du To investigate the efficacy, safety and tolerability of saxagliptin compared with with saxagliptincompared of andtolerability safety efficacy, Toinvestigatethe 488 patients (1:1)to wererandomized SMART was a 24-week, multicentre, randomized, parallel-group, multicentre, parallel-group, randomized, wasa24-week, SMART ≥ 1 dose of study medication. study doseof 1 n was non-inferior to acarbose forglycaemicn wasnon-inferiortoacarbose oportion of patients achieving HbA1c<7.0%, ofpatients HbA1c<7.0%, achieving oportion adverse events (GI events and AEs), of adverse theproportion saxagliptin or viaacarbose saxagliptin acentral Safety and tolerability were also assessed andtolerability Safety ≥ 1 pre- and 1 pre-

≥ 1 post- This articleis protectedbycopyright. Allrightsreserved. Accepted Article specifically in the Chinese population. inChinese specifically the trials forclinical the need highlight T2DM with patients Chinese of characteristics registration number: registration trial Clinical might forclinicalassociated suggestingfewer GIAEs, bepreferential withitpractice. Conclusion: acarbose. those receiving compared withoutGIAEs with HbA1c<7.0% achieved saxagliptin whoreceived ofpatients number agreater and acarbose with compared saxagliptin treatment However, achieving fewer HbA1c<7.0%. GIAEs were reportedwith groups with 0.13]%) [–0.22, –0.04 interval]: confidence mellitusprediabetes. (T2DM)and50.1%having 2diabetes type 11.6%ofadultshaving estimated an epidemic with is inChina, Diabetes Introduction number of Chinese individuals are overweight (body mass index [BMI] massindex (body overweight are individuals ofChinese number without significantly increasing the riskofhypoglycaemia. without significantly postprandial levels, fa (PPG) HbA1c, developing T2DM at a lower body mass index (BMI; 25 vs 30 kg/m 25vs30 index (BMI; mass body a lower at T2DM developing to more susceptible are people Chinese Furthermore, fordiabetes. factor risk important beingan adiposity with mainthe population, ofT2DMinthe causes Chinese considered antihyperglycaemic (OADs). drugs glucosidase inhibitors (including reduce the digestion of in of reduce the digestion In China, metformin is the recommended first-line therapy for patients with T2D with forpatients therapy first-line recommended the is metformin In China, (BMI in the early stages of diabetes compared withWesterncompared populations. ofdiabetes stages early in the withT2DMha people addition,Chinese ≥ 30 kg/m Both therapies had similar efficacy profiles. However saxagliptin was saxagliptin profiles.However efficacy similar had therapies Both 2 ), 1, 3-6 and there has been an increased prevalence of central obesity. central of prevalence increased been an has andthere acarbose) are the most-widely used class of second-line oral ofsecond-line usedclass most-widely are the acarbose) 10, 11 10, NCT02243176, clinicaltrials.gov. NCT02243176, 10 They act in theborder ofthesmallintestinein to brush They act

ve been observedto have worse the first half of the small intestine. small half ofthe the first sting plasma glucose (FPG) and body glucose(FPG) weight, plasma and body sting similar proportionsofpatientsin both 1, 2 Poor diet and lack of physical activity are activity ofphysical dietandlack Poor 10-13 However, gastrointestinal (GI) 9 These differences in the the in differences These 2 ). 1, 3-7

≥ β 25 kg/m 25 -cell deterioration -cell deterioration An increasing An 12 Thisreduces 11 2 , and ) orobese 3, 8 In α - This articleis protectedbycopyright. Allrightsreserved. Accepted Article visit.randomization an FPG and and11.0% 7.0 and anHbA1c between screening, at and 11.0% between 7.5 HbA1c an with weeks 8 foratleast label) the on specified maximumdose more thanthe toleratedbut not maximally dose, or individually mg/day included ( patients The study at sitesinChina. thatwasconducted monotherapy 35 metformin controlled inadequately with with T2DM with acarbose inpatients saxagliptin comparing IVstudy Phase open-label mu a24-week, (NCT02243176) was The SMARTstudy participants andStudy design Materials andMethods inadequately controlled with metformin monotherapy. T2DM with patients inChinese with acarbose compared of saxagliptin tolerability and safety efficacy, investigatedthe Study TheSMART some patients. for acarbose and abdominal pain in pain abdominal and side effects are common, with occurring in occurring withflatulence common, are effects side neutral. weight- being body as aswell risk ofhypoglycaemia, with alow control improve glycaemic (DPP-4) peptidase-4 is adipeptidyl Saxagliptin secretion and suppression of glucagon of secretion. insulin secretionand suppression (GIP), insulinotropic peptide glucose-dependent tolerability, profilesimilar to efficacy a have DPP-4 inhibitors 15-18 DPP-4 inhibitors prevent the inactivation of glucagon-like peptide-1 (GLP-1)and glucagon-like peptide-1 theinactivation of prevent DPP-4 inhibitors 14,20, 15, 21 and so may be a more attractive choice of second-line therapy thanmore second-linemaya therapy of andsobeattractivechoice ≥ 1/100 to<1/10patients. ≥ 18 years) treated with stable metformin monotherapy ( monotherapy metformin stable treated with 18 years) inhibitor that has been demonstrated to inhibitorhas beendemonstrated that 14 resulting in increased glucose-dependent increased glucose-dependent resulting in

α -glucosidaseinhibitors but with betterGI ≥ 1/10 1/10 patients, and diarrhoea and GI 19 lticentre, randomized, parallel-group, Previous data have shownthat Previousdata ≤ 13.3 mmol/L at the pre- mmol/Lat the 13.3 ≥ 1500 This articleis protectedbycopyright. Allrightsreserved. Accepted Article is design study if appropriate.The mg, monitor to 100 the dose 7to foradverse andtotitrate onDay (AEs) consultation events and (<8.0% HbA1c voice/central (interactive system randomization a by randomized were Patients of thestudy. end tothe lead-in period implemented from the was that plan exercise and a diet received allpatients In addition, adjusted. not be could otherwise but inresponsemetformin tohypoglycaemia, could bereduced dose study; themetformin throughout of and regimen existing dose ontheir continued also All patients if appropriate. mgdaily three-times to100 titrated randomization,then and following 7 days injury or secondary forms of diabetes; history diabetes;history forms of or secondary injury oftypecriteria Exclusion diagnosis included: acarbose (Glucobay saxagliptin (Onglyza saxagliptin either toreceive system randomization central (1:1)viaa randomized were Patients idiopathic or of chronicpancreatitis or ahistory (NYHA) class III significa orIV; Heart Association New York as defined failure heart congestive screening; to prior 4 weeks or within at screening disease renal end-stage or impairment renal moderate/severe screening; mmol/L priorto weeks at screeningor within 4 >4.5 triglycerides before8 weeks screening; inthe days consecutive more metformin than7 for than other agent anti-hyperglycaemic any of hypersensitivityhistory reaction either to receptor GLP-1 ofscreening; agonists within inhibitor 1 year or treatment DPP-4 any with or hyperosmolar coma (including ketoacidosis ® ≥ ® , AG, Leverkusen, Germany) 50 mg three-times daily for thefirst daily mg three-times 50 Germany) Leverkusen, AG, , Bayer 8.0%). Those receivingacarbose additional telephone Those an had 8.0%). , AstraZeneca, Möndal, Sweden) 5 mg once daily for 24 weeks or for 24weeks mg daily once 5 Möndal, Sweden) , AstraZeneca, nt cardiovascular history within 3 months priortoscreening; within3 history cardiovascular nt shown in shown Figure Supplementary 1. a DPP-4 inhibitor or acarbose; treatment with oracarbose;treatment inhibitor with a DPP-4 1 diabetes, diabetes resulting from pancreatic resulting diabetes, diabetes 1 of acute metabolicof acutecomp ) within 6 monthswithin ofscreening; ) 6 previous web responsesystem),toaccording stratified acute pancreatitis or gastrointestinal disease. disease. orgastrointestinal acute pancreatitis lications of diabetes lications ofdiabetes This articleis protectedbycopyright. Allrightsreserved. Accepted Article frequency ofAEsand serious AE frequency a ofsaxagliptin andtolerability The safety (homeostasis model assessment- PPG, PPG, 2-h baselineinFPG, from change AEs, andthe GI Week 24without at response glycaemic AEs, GI any with ofpatients the proportion 24, atWeek <7.0%) asHbA1c (defined response glycaemic therapeutic a achieving patients of proportion the outcomes were efficacy Secondary acarbose. with compared saxagliptin 24with Week at HbA1c in frombaseline change absolute was outcome efficacy The primary outcomes Study objectives and informedwritten consent. provided all patients and at eachsite committee theinstitutionalethics orindependent by protocolwasboard approved review study The requirements. regulatory applicable and all Helsinki, of theDeclaration in their origin on Conference Harmonization–Good wasth inaccordancewith performed The study confirmation scheduled. so thatthetestcould was alertedbe investigator the criteria discontinuationmet oftheFPG Ifapatient visits. one pre-randomization toWeek (randomization thestudy course of the pa efficacy wereblindedtokey investigators a wasopenlabel, measurement second thestudy week).confirmed within Although by 1 mmol/L at(Week 12; Visit 7 orifFPGwas>12.2 measurement 1week) within second mmol/Lat a confirmed Patients by discontinued6(Week4; Visit was >13.3 were ifFPG β -cell function and body weight at Week 24. body atWeek24. weight -celland function β s (SAEs), as well as monitoring of vital signs, clinical vital clinical signs, monitoring of aswell (SAEs), s ). Clinical Practice guidelines, ethical principles that have that ethicalprinciples guidelines, Clinical Practice nd acarbose were evaluated according to the the proportion of patients achieving therapeutic achievingtherapeutic the proportionofpatients rameters (HbA1c, FPG and2-hourPPG)during FPG rameters (HbA1c, e clinical trial protocol, the International protocol, trial e clinical and thescreening of exception the 24) with β -cell function was measured HOMA- by -cell was function β

This articleis protectedbycopyright. Allrightsreserved. Accepted Article who received at least one dose of study medication at one andhad baselineand received leastone at ofstudy who oneleast dose patients all included which set(FAS), analysis inthe full analysed was outcome The primary acarbose, and respectively.rates withsaxagliptin of12.5and25% assuming AE AEs acarbose, between GI ofGI and saxagliptin incidence the in difference to 85%detect power a approximately provide wouldalso population 480 A study of devia standard the andthat treatments between same trueeffectswould bethe thatthe rate This assumed drop-out of20%. expected upon an atendpoint a one-sided power toestablishof non-inferiority saxaglip enro tobe were planned patients A totalof480 analysis Statistical evaluated. acarbose wasalso with compared with saxagliptin at 24 (EQ-5D) 5 Dimensions Week of Life – Quality in theEuropean baselinechange from counts. tabletThe based onreturned assessed was Compliance orbehaviour. in consciousness impairment severe to due assistance external events requiring assymptomatic defined events were Severe hypoglycaemic asympt oran forexternal assistance, need mmol/L no and level <2.9 symptomaticevent bloodglucose a aswith either were defined events atin either treatment in that least2%ofpatients group.Hypoglycaemic those occurred as defined AEswere Common group. treatment accordingto frequency by and sorted term, Activitiesfor Regulatory (MedDRA) 18.0, summa Medical according tothe andDictionary AEs were coded parameters. haematology chemistry α level of 0.025, with a non-inferiority margin with andbased a levelnon-inferiority 0.4% of0.025, of omaticmeasurementblood glucose mmol/L. <2.9 tin compared with acarbose for the primary theprimary withacarbose for tin compared tion (SD) for the change in HbA1c was 1.2%. was HbA1c in (SD)change for the tion lled (240 patients per study arm) to provide 90% arm) toprovide (240patientsperstudy lled rized by system by rized organclassand preferred This articleis protectedbycopyright. Allrightsreserved. Accepted Article took at least one dose of study treatment. Analyses for safety outcomes were summarized summarized outcomes were forsafety treatment.study Analyses one doseof took atleast who all included patients set which analysis safety the in evaluated outcomes were Safety CIwere esti associated 95% risk and or (<8.0% at baseline HbA1c with test Haenszel Cochran–Mantel– using the analysed were variables categorical as recorded outcomes andHOMA- PPG in2-h from baseline outcome (change primary ofthe analysis for the usedsimilar sensitivity that anweight)ANCOVA to or and body FPG in baseline from (change outcome for theprimary used that to similar analysis MMRM using a outcomeseither analysed variableswereas continuous efficacy recorded Secondary premature andpoorcompliance. discontinuation outcome mightaffect study the that deviations all FAS that oftheFAS asubset was included set analysis Theanalysis. PP setasasupportive analysis (PP) in the per-protocol investigated also outcome was primary The covariate. a as at baseline value and HbA1c effect fixed missingobservationcarried for forward valu conducted theoutcome primary on ofcovariance usinganalysis withlast (ANCOVA) wasalso analysis Asensitivity baselineerror[SE])betweengroups. from (standard change mean the leastsquares as thedifference eachin baselinetreatment within aswell group, confidenceintervals (CIs) were presentedat each visit forthe leastmean squares change from missing effic of observedimputation without and time-by-treatment-group as measurement interaction.Dataanalysed baseline were for treatmentgroup,time, terms including model repeated-measures analysis (MMRM) mixed- usinga analysed measurement. was outcome efficacyThe primary post-baseline mated using the Mantel–Haenszel method.mated usingthe Mantel–Haenszel es. The ANCOVAincludedtreatmentes. The groupasa acy estimates data. Point 2-sided95% acy and . The most frequent protocol deviations were most deviations frequentprotocol The . patients who did not have significant protocol significantprotocol nothave did patients who ≥ 8.0%) as a stratification factor. The relative relative The factor. asastratification 8.0%) β ). Secondary efficacy efficacy ). Secondary This articleis protectedbycopyright. Allrightsreserved. Accepted Article n=15 [6.1%]), most frequently because of a patient’s decision to discontinue (Figure 1). (Figure 1). to discontinue decision because ofapatient’s most [6.1%]), frequently n=15 Similar in ofpatientsdiscontinued numbers bot respectively. and 204pa therefore included 219 set analysis PP groupand 39 saxagliptin in group the acarbose ha acarbose groups and 243 patientsin saxagliptin the 238and included meaning thatFAS measure, the efficacy post-baseline one atleast not have gr saxagliptin inthe additionalThree patients acarbose group. and 244patientsin the saxagliptingroup setinthe includedpatients analysis 241 the safety medication. Therefore, study groupdidnottake any in thesaxagliptin patients oracarbose Three (n=244) (n=244). (1:1)toreceive randomized saxagliptin patients were (Figure 1).Following screening, 488 2015 and29September 2014 enrolled 24September between patientswere Overall, 689 Disposition Results of the theoutcome.sensitivity analysis used for primary 24 (patientanalys 0 andWeek reportedoutcome at Week both questionnaires who completed EQ-5D inpatientsFAS the included in analysed index EQ-5D were as the frombaseline theChange item ineach aswell individual of EQ-5D categorical variables. statistics va using descriptive forcontinuous

riables or frequency count riables orfrequency oup and one patient in the acarbose group did acarbose group patientand one in the oup h groups (saxagliptin,h groups n=14 [5.7%]; acarbose, is set), using an ANCOVA similarset),an ANCOVA is using that to tients in the saxagliptin and acarbose saxagliptin tients in groups, the d significant protocol deviations, and the d significantdeviations, protocol , respectively. patients in Nineteen the s and percentagess and for This articleis protectedbycopyright. Allrightsreserved. Accepted Article and 26.3kg/m 26.4 BMIwas mean The acarbose (Tablereceiving 1). the with group compared saxagliptin group receiving the in which washigher of2-h PPG exception the with treatment groups, between wellbalanced FAS were generally the for characteristics and baseline Demographic demographics characteristics and Baseline MMRM analysis ofthePP populat as the to 0.13%),aswell CI:–0.22% [0.09]%; 95% –0.04 mean [SE]difference: squares least respectively; and acarbose, with saxagliptin [0.06]% and –0.76 [0.06]% Week 24: –0.81 (leastsquare (ANCOVA)in theFAS analysis the results wassupportedby the theendpoint.This of sensitivity to acarbosefor primary margin than non-inferiority 0.4%but greater of CIwas belowthe 95% limit ofthe The upper CI:–0.22%to0.13%. was –0.04(0.09)%;95% squaresmean (SE)difference analys (MMRM least (Table 2).The group,respectively acarbose and group inthe0.78 (0.06)% saxagliptin of–0.82(0.06)%and– in HbA1c 24 (MMRM analysis) (SE) changefrom toWeek baseline mean (Week least squares 24).;the study end ofthe tothe up continuing 2) withreductions le HbA1c reduced acarbose and Both saxagliptin outcome efficacy Primary Efficacy outcomes T2DM. with population a expected in wouldbe groups,as in both included individuals and obese 2 in the saxagliptin and acarbose groups, respectively, with both overweight withbothoverweight respectively, acarbose groups, and saxagliptin in the ion (Supplementary table1). (Supplementary ion s mean [SE] change from baseline to is) between the groups (saxagliptin – acarbose) 0, indicating the non-inferiority ofsaxagliptin indicatingthenon-inferiority 0, vels rapidly Week4oftreatment; (by velsFigure rapidly This articleis protectedbycopyright. Allrightsreserved. Accepted Article (occurring in with treated acarbose. patients (20.1%) and 49 saxagliptin with treated patients (1.2%) three in occurring wereGIevents, AEs related occurringtreatment-most frequently in each(Tablegroup, respectively 3).The patients and50(20.5%) occurringacarbose in8(3.3%) group, compared withthe group saxagliptin lowerinthe AEswas of treatment-related Thefrequency days, respectively. (24.64) to saxagl meanof exposure duration The (SD) Safety andtolerability and 99.9%forpatients100.2% acarbose,and receiving respectively. saxagliptin treatmenthigh inmean treatmentcompliance was groups; treatment both was to compliance addition, Table (Supplementary In 1). in the PPpopulation observed also endpoints were these with compared acarbose with wasobserved loss 2-h PPGandHOMA- FPG, in Week 24 tr significant between There wasno difference AEs. GI responsewithout glycaemic therapeutic a achieved respectively, acarbose, and saxagliptin receiving patients and66(28.8%) 85(37.0%)patients Overall, p<0.0001). riskratio=0.19, [26.0%]; acarbose, 53/204 [5.0%]; 11/219 population (saxagliptin, the inPP GI AEs (risk observed AEs Thislowerriskof was also ratio=0.22, p<0.0001). (5.5%) patients receivingand60(24. saxagliptin IntheFAS,13 <7.0%). response(HbA1c glycaemic atherapeutic achieved had acarbose patients receiving (41.5%) and95 patients receivingsaxagliptin 88(38.3%) At Week24, Secondary efficacy outcomes ≥ 2% of the safety analysis set) were abdominal distension and flatulence which and flatulence which distension abdominal set)were theanalysis 2% of safety Commonly occurringtreatment-related AEs β 2),however,significantly greater(Table weight iptin and acarbose was 165.5 (21.93) and 163.2 (21.93)and163.2 was165.5 acarbose and iptin eatment groups for changeeatmentto groupsfor frombaseline saxagliptin (p=0.0078). Consistent resultsfor (p=0.0078). saxagliptin 7%) patients receiving acarbose reported GI patientsacarbose reported receivingGI 7%) This articleis protectedbycopyright. Allrightsreserved. Accepted Article GI AEs there was no significant difference be difference significant GI AEstherewasno AEs. GI Despiteglycaemic this response without a therapeutic achieved saxagliptin received who ofpatients number agreater and acarbose, However, with fewer compared (HbA1c <7.0%). werereportedsaxagliptin GIAEs with inbothtreatmentof patients group proportions the effectsbetween of saxagliptin an difference nosignificant there was Furthermore, control. glycaemic inadequate with patients metforminin add-on to as for reductioninHbA1c, was non-inferiortoacarbose saxagliptin patients in typical of apopulation This study Discussion andacarbose. saxagliptin difference betweenno overallapparent Table3),with treatment (Supplementary groups inboth weresimilar itemsthe EQ-5D all of from in baseline changes the At Week24, outcomes Patient reported groups; 3(1.2%)patientsreceivingsaxaglipti was inboth lowandsimilar ofhypoglycaemia Table2).Theincidence (Supplementary termswith saxagliptinof haematology, clinical in oracarbose chemistry vitalsigns observations nounexpected Therewere of saxagliptin. to discontinuation did notlead SAE This symptomatic treatment. Theoedema following 106days. resolved for the drug taking lower limbsafter both relatedexperiencing of oedema moderate potentially the treatment, to an SAE saxagliptinhad patienttreated with ineach group.One patient discontinuing inbothgroupswithone(0.4%) low also was AEsleading to discontinuation of The incidence saxagliptin. treatedwith respectively, zero patients, (0.8%) and patients, 23 (9.4%)and17(7.0%) occurred in d acarbose on FPG, 2-h PPG or 2-hHOMA- PPG d acarbose FPG, on with T2DMinthat China demonstrated n andn 4(1.6%)patients receivingacarbose. tween treatment groups in thechange from respectively, treated with acarbose, and two two and acarbose, treatedwith respectively, s achieved a therapeutic glycaemic response glycaemic therapeutic a s achieved difference in the proportion of patients with proportion ofpatients with in the difference β . Similar This articleis protectedbycopyright. Allrightsreserved. Accepted Article The current Chinese treatment guidelines for T2DM The currentguidelines treatment Chinese study. in this was observed 24weeks over hypoglycaemia of acarboseand alowincidence safety findings with saxagliptin or unexpected addition, no there were In pain/discomfort. including items, EQ-5D in Week 24 to baseline mg/day is currently higher than the daily cost of saxagliptin. cost of thedaily higherthan is currently mg/day of300 a dose at acarbose cost of the daily In addition, theclinicaltrialbe related to setting. between treatment armswas obser compliance regimens treatment burdensome forless a preference have aspatients important is This acarbose. required. The redu useofsaxagliptin,therefore, mg if three-times increasedto100 daily, mg three-timesdaily, at 50 initiated generally the ofsaxaglipti with benefits theoretical use low riskofAEs. witha efficacy antihyperglycemic provide shownto been andhas withother OADs, therapy andas combination monotherapy as Chinesein the population studied has been Saxagliptin individual patient. facilitating selectionof thereby consideration, for beidentified to anddisadvantages advantages potentialdifferences, any enable as they choices treatment ofthese studies Head-to-head inhibitor. insulin orDPP-4 secretagogue, inhibitor, alpha-glucosidase an be could whichcurrent additiontreatment, recommendthe a second-line guideline the of metformin control with maintaincan glycaemic longer no oncepatients However, metformin. initiatedexercise are on and controlled with poorly diet with T2DM contraindicated, patients 23 , which has also been observed to improve compliance. alsobeenobservedtoimprove , whichhas 15-18, 20-22 In addition to its good tolerability, there are ofother a number there toits goodtolerability, Inaddition n. It is given once-daily, whereasacarbose is once-daily, is given n. It the optimum the second-linetreatment foreach ved in the SMART study, however, this may in SMARTstudy, however, this ved the ces the pill burden for patients compared with with compared forpatients burden pill the ces are, therefore, important for clinical practice practice for clinical important are,therefore, 11 recommendunless that, 26

24, 25 No differencein This articleis protectedbycopyright. Allrightsreserved. Accepted Article rice) resulting in larger PPG excursionspeople. inAsian inlargerrice) PPG resulting produced physiologically inresponsetoeating.produced physiologically inhibiting actthe degradatioDPP-4 inhibitors by PPG levels a have larger impactPPG hypergly on that suggested it hasbeen as acarbose, manner to a similar in Asianpatients for beneficial may be andthis PPG on act predominantly they of action mechanism totheir Owing cont which of release,both glucagon and inhibit safety profile. safety inits significantchanges any 2 without years, beyond maintained of saxagliptin is efficacy in non-Chinese longer-term studies However, or safety data for efficacy provide long-term does not andso 24weeks duration only was the study Inaddition, assessmentat 24weeks. theEQ-5D mighton them registeredmedication), have patients not with symptomatic (beforetreatment treatment initiation following weeks the occurredfirst tohave during likely are GIAEs most as 24 and, and Week at baseline evaluated was only EQ-5D results. The the considering mind when limitationsin that has bekept of should anumber This study metformin.following failure of be may that it AEs.Thissuggests fewer GI and effects onPPG their profiles, including that demonstrated thisstudy In conclusion, 31 This might be related to the Asian diet, which includes food with high glycaemic load(e.g. withhigh food mightdiet, whichincludes berelated This to theAsian 32, 33

saxagliptin andacarbose have similar efficacy preferable to acarbose forsecond-line use to preferable β caemia than FPG levels in Asian patients. in caemia levels FPG than the use of saxagliptin compared with acarbose. acarbose. with compared ofsaxagliptin use the -cell function, saxagliptin but isassociated with populations have shown that the glycaemic the that shown have populations 27, 28 ribute to a reduction in blood glucose levels. glucoselevels. inblood a reduction ributeto n of the incretins GLP-1 and GIP, which are which GLP-1 andGIP, theincretins of n The incretins stimulate insulin secretion secretion insulin stimulate Theincretins 31

12, 29- 18, This articleis protectedbycopyright. Allrightsreserved. Accepted Article AstraZeneca; and been a speaker for Novo Nordisk, Eli Lilly, AstraZeneca, SanofiAventis AstraZeneca, EliLilly, NovoNordisk, aspeaker for AstraZeneca; andbeen and Eli Nordisk, Lilly, for Novo boards advisory attended and Bayer. LL has for EliLilly andbeena speaker FXhasreceived andNordisk. lecturefeesfrom EliLilly, Nordisk Novo fees from Novo receivedlecture CX has and EliLilly. Aventis, Bayer Sanofi AstraZeneca, HF and Bayer. Aventis Sanofi AstraZeneca, XW has these for been aspeaker companies. trialsclinical the in WLhasparticipated and Bayer. Sanofi Aventis, AstraZeneca, a speaker EliLilly, NovoNordisk, and been for fromNovartis; research grants received and AstraZeneca; Nordisk, EliLilly, for Novo boards JD hasreceivedlecture feesfrom NovoNord Interest Conflicts of writingofthereport.data,and the design, th in study the involved AstraZeneca was fundingRole ofthe source Ltd. Springer Healthcare Communications, of inScience Jones Alexander by andprovided China Ed AstraZeneca. supported by was This study Acknowledgements manuscript. approval ofthe review andfinal andpreparation, data collection, trial design, operation, wereinvolved inthe study All authors manuscript and preparation. analysis trial operation,data design, study ledthe Yiming Mu Author contributions of Novo Nordisk, Eli andSanofiAventis; Nordisk, has Lilly of Novo had been a speaker for hadbeenaspeaker Novo Nordisk, been a speaker for Novo Nordisk, forNovo EliLilly, been aspeaker isk and Sanofi. YM has attended advisory advisory isk YMhasattended and Sanofi. itorial assistance was funded by AstraZeneca AstraZeneca fundedby assistance was itorial e andcollection, analysis interpretation of This articleis protectedbycopyright. Allrightsreserved. Accepted Article 5. The burden of overweight andob overweight of Theburden 5. in China.Urbanisation et andhealth Carltonal. EJ, LiangS, GongP, 379: 843-852. 4. in China. diabetes 2 oftype X, Jia W.Causes Lin MaRC, Endocrinol. 3. adults. inChinese ofdiabetes andcontrol al. Prevalence et Y,WangL,HeJ, Xu 2013; 310:948-959. 2. forapersonal solution societal a inChina: Ning G.Diabetes Y, Zhang ChanJC, challenge. 1. References Sanofi Aventis, MSD. Takeda, Novartis, Ingelheim, Boehringer Nordisk, clinical in trials the participated of Nova LShas from Nordisk. and lecture Novo grants fees FB hasreceivedresearch and Bayer. 9. Kong AP, Xu G, Brown N, So WY, Ma RC, Chan JC. Diabetes and its comorbidities-andits JC.Diabetes SoChan KongAP, WY, Ma RC, N, Xu G,Brown meets West. -where East 9. surveys. two cross-sectional obesity changes ofoverweight, inthe prevalence LuH,etal. LiuY, Ten-year Hou X, -comparison of urban Shanghai,1998-2007 in adults amongtheChinese obesity central and 8. Fail. Heidenreich PA, Albert NM, Allen LA,et Heart Association. the American from statement a policy States: United in the 7. children in Chinese dynamics WangBM.Overweight H,Popkin Gordon-Larsen P, and adults. 6. 191-196. in Chinese patients. patients. in Chinese of diabetes treatment the in ofacarbose andefficacy Safety XM. JC, Mao HeK,Shi 10. M, eds. Managing Diabetes. London: Springer Healthcare Ltd. 2012; 61–89. 2012; 61–89. Ltd. SpringerHealthcare eds.M, Diabetes. London: Managing Evans InVoraJ, forDiabetes. Agents S. Non-Insulin Shankarnarayan G, Sreemantula 12. 2013: – 2013. type in China 2diabetes Association. Diabetes ofMedicalCare Standards 11. 2013; 6: 606-619. 6: 606-619. 2013; Lancet Diabetes Endocrinol. Obes Rev. Obes 2014; 2: 980-991. 2014; 2:980-991. Ther ClinTher RiskManag. 2014; 15 Suppl 1: 37-48. 1:37-48. 15Suppl 2014; Nat Rev Endocrinol. Nat Rev BMC PublicHealth. 2014; 2: 969-979. rtis and Takeda; has been aNovo has speakerTakeda; for and rtis esity in the Asia-Pacific region. region. intheAsia-Pacific esity 2014; 10: 505-511. 2014; 2013; 9: 537-547. 9:537-547. 2013; 2013; 13: 1064. 2013;13:1064. al. Forecasting the impact of heart failure failure ofheart impact the al. Forecasting Lancet Diabetes Lancet Diabetes Obes Rev. Obes Lancet. Circ Heart Heart Circ 2007; 8: 2007;8: 2012; 2012; Jama.

This articleis protectedbycopyright. Allrightsreserved. Accepted Article Adherence. 2diabetes. type with patients among adherence andlikely to pay, willingness burden ondosingpreferences, pill JC,etal.Effect of S,Yang HauberAB,Han 23. Diabetes MetabRes Rev. trial. controlled randomized a mellitus: 2 diabetes type with Asian patients in drug-naive saxagliptin of and safety I,ZhaoJ.Efficacy W,TouC,Gause-Nilsson PanCY,Yang 22. metformin. controlled with inadequately 2diabetes type aged in acarbose than control metabolic in effective more variability inglycaemic similar et is YM, al.Saxagliptin Lin S,Chen WangMM, 21. trial. controlled a randomized mellitus: 2diabetes with type metformin people inAsian to added TouC,ZhaoJ,Gause-Nil YangW,PanCY, 20. 1705. in 2diabetes. inhibitors type peptidase-4 dipeptidyl agonists and receptor peptide-1 glucagon-like system: Theincretin NauckMA. DruckerDJ, 19. diabetes. modulating postprandial glucagon and C-peptide J,Hirshberg Lu SjostrandM,Iqbal N, 18. Diabetes Care. Association(ADA) andtheEu Diabetes ofthe American statement position approach: a patient-centered diabetes: 2 type in of RM, BuseJB,etal.Management Bergenstal Inzucchi SE, 17. assessing cardiovascular data. mellitus: forthetreatment2diabetes Cobble Saxagliptin FrederichR. oftype ME, 16. week, post-marketing SUNSHINEstudy. addedmetformin to inChinese patients withtype 2diabetesmellitus: results from the 24- et GuW,Liangal.Efficacy L,WangS, 15. Diabet Med. trial. randomized double-blind, a24-week, 2diabetes: inwithType patients monotherapy acarbose and of etal.Comparison JP, W,Barona PanC,Yang 14. e79421. inhibitors on patients between Asian and Caucasian. and Caucasian. Asian between 2 diabetes patients type inhibitors on alpha Y, Ji ofthe glucosidase Luo efficacy L.Comparisons HanX, CaiX, 13. Diabetes Res Clin Pract. Clin Res Diabetes Diabetes Res Clin Pract. Res Diabetes 2013; 7: 937-949. 2008; 25: 435-441. 2008;25:435-441. 2012; 35: 1364-1379. 35: 1364-1379. 2012; 2012; 28: 268-275. 2012;28:268-275. Diabetes Res ClinPract. Diabetes 2011; 94: 2011; 217-224. Cardiovasc Diabetol. Cardiovasc ropean AssociationDiabetes (EASD). fortheStudyof 2014; 105: 185-191. 2014; 105:185-191. J Diabetes. B. Saxagliptin improves glycemic control by glycemiccontrolB. Saxagliptinimproves by and safety of saxagliptin monotherapy or monotherapy saxagliptin ofand safety levels in Chinese patients with type 2 withtype patients inChinese levels 2012; 11: 6. 6. 11: 2012; sson I. Efficacy and safety of saxagliptin ofsaxagliptin and safety I.Efficacy sson 2015; 108: e67-70. e67-70. 108: 2015; 2015: 2015: Lancet. PLoS One. PLoS 2006; 368: 1696- 2006; Patient Prefer Prefer Patient 2013; 8: 2013; This articleis protectedbycopyright. Allrightsreserved. Accepted Article in patients with type 2 diabetes mellitus. mellitus. 2diabetes type with in patients outcomes cardiovascular and al. Saxagliptin et BM, BhattDL,Braunwald E, Scirica 32. and Caucasians. postprandial glycaemiainAsians MannWilliams JI.Comparison of VennBS, SM, 31. Diabetes MetabRes Rev. diabetic pa hyperglycaemia 2 type in Asian excess glucose to et ofpostprandial al. Contribution WangJS,TuST, LeeIT, 30. pathophysiology. riskfactors,and epidemiology, inAsia: al. Diabetes JiaW, et V, ChanJC,Malik 29. agents? as newantidiabetic work dothey inhibitors: how DemuthPospisilikHU, IV McIntoshCH, peptidase JA,PedersonR.Dipeptidyl 28. peptidase-4 (DPP-4). ofdipeptidyl- ofaction ofinhibitors Gallwitz NA, B.Mechanism Thornberry 27. l. Accessed. http://www.zjpi.gov.cn/main/html/2014/CT10046/dd72e4240a7e4c1da820a849642f7b37.htm foracarbose. pricelist drug Official Price Bureau. 26. Adherence. Improving Patient AM. Delamater 25. Care. fordiabetes. medications of adherence with review Asystematic Cramer JA. 24. Age, years (mean [SD]) [SD]) (mean years Age, saxagliptin in drug-naive and metformin-treated patients with Type 2diabetes. metformin-treatedpatients withType indrug-naiveand saxagliptin J, Rosenstock Gross JL, Aguilar-Salinas C, etal.Long-term 4-year safety of 33. Saxagliptin (n=238) Acarbose (n=243) Overall (n=481) (n=481) Overall Acarbose (n=243) Saxagliptin (n=238) 1. Table Tables Figure 2. Figure 1. Figure legends 2013; 30:1472-1476. 2004; 27: 2004; 1218-1224. Demographic and Demographic baseline charac Change Change from baselineover inHbA1c time study throughout the Patientdisposition Diabet Med. Jama. Best PractResClinEndocrinolMetab. 2009; 301:2129-2140. 2011; 27: 79-84. 2011;27:79-84. 2010; 27: 1205-1208. 27: 1205-1208. 2010; 47(05) 65(08) 55.6(10.69) 56.5(10.81) (10.51) 54.7 N Engl J Med. N Engl teristics (full analysis set) (fullanalysis teristics tients using continuous glucose monitoring. glucose using continuous tients Clin Diabetes. 2013; 369: 1317-1326. 369: 1317-1326. 2013; Regul Pept. 2009; 23: 479-486. 2009; 23:479-486. 2006; 24: 71–77. 71–77. 2006;24: 2005; 128: 159-165. 128: 159-165. 2005; Diabet Med. Diabet Diabetes Diabetes

This articleis protectedbycopyright. Allrightsreserved. Accepted Article mellitus. 2diabetes type T2DM, deviation; standard SD, mass plasma 2-hourpostprandial fasting glucose; index;BMI, body 2-h PPG, glucose; FPG, [SD]) [SD]) PPG,mmol/L(mean 2-h FPG,mmol/L (mean [SD]) [SD]) %(mean HbA1c, [SD]) years (Meandiagnosis, ofT2DM Duration BMI, kg/m Weight, kg(mean[SD]) [%])Female (n *n=237; *n=237; † n=480; 2 (mean[SD]) ‡ n=235; § n=478. 33(26) 7. 1.7 72.9(12.43) 72.6(12.27) (12.61)* 73.3 11.17 (2.82) 11.17 64(.7* 63(.9 26.3 (3.48) 26.3(3.49) (3.47)*26.4 9.01 (2.14) 9.01 .3(.5 81 08) 8.20(0.83) 5.2(4.58) 8.16(0.81) 5.3(4.76) 8.23 (0.85) 5.1 (4.40)* 91 (38.2) 105 (43.2) 196 (40.7) 196 105(43.2) 91 (38.2) ‡ ‡ .1(.5 8.90(2.04) 8.81(1.95) 02 28) 10.70(2.89) 10.25(2.89) † § †

§ †

This articleis protectedbycopyright. Allrightsreserved. Accepted Article Difference (SE) –0.04 (0.09) –0.04 –0.78(0.06) 243/229 7.34(1.04) 8.16(0.81) –0.82(0.06) / 238 230 7.38(1.03) 8.23(0.85) Difference (SE) means (SE) change LS [SD]) (Mean Week 24,% [SD]) (Mean Baseline, % Weeks BL / n at 24 (MMRM analysis) HbA1c Saxagliptin postprandial glucose, HOMA- 2. Table (n=238) Acarbose (n=243) p value 0.0078 0.0078 0.69(0.26) 0.3739 –2.05(0.18) 243/226 70.4(12.6) 72.6(12.3) –1.36(0.18) 7.48(8.407) / 237 227 71.6(12.5) 73.3(12.6) SE, standard error. 0.2248 mixed-modelMMRM, repeated-measures; standard deviation; SD, 13.08(5.958) HOMA- fasting plasma glucose; 2-hourpostprandialANCOVA, glucose; 2-h PPG, p value 20.56(5.932) 0.30(0.249) 68.3 (66.7) Difference (SE) 55.6(59.5) means (SE) change LS [SD]) (Mean Week 24,kg 0.8915 [SD]) (Mean Baseline, kg –1.07(0.174) 75.6(118.5) Weeks BL / n at 24 55.4(54.5) analysis) (MMRM weight Body p value 9.38(2.75) –0.77(0.176) 0.02(0.18) Difference (SE) 10.25(2.89) means (SE) change LS mU/mmol Week (Mean[SD]) 24, 0.6236 10.04(2.79) –1.01(0.13) mU/mmolBaseline, (Mean[SD]) 243/229 11.17(2.82) Weeks BL /24 n at HOMA- 7.76(1.92) p value –0.99(0.13) 8.81(1.95) Difference (SE) / 235 227 means (SE) change LS mmol/L (Mean[SD]) Week 24, 7.87(1.98) mmol/LBaseline, (Mean[SD]) 9.01(2.14) Weeks BL /24 n at analysis) PPG(ANCOVA 2-h p value Difference (SE) means (SE) change LS mmol/L (Mean[SD]) Week 24, mmol/LBaseline, (Mean[SD]) Weeks BL / n at 24 FPG analysis) (MMRM p value Change from baseline to Week 24 in HbA1c, fasting plasma glucose, 2-hour 2-hour plasma fasting glucose, 24inHbA1c, to fromWeek baseline Change β (ANCOVA (ANCOVA analysis)

β and body weight (full analysis set) set) (fullanalysis weight andbody β , homeostasis model assessment- analysis of covariance; BL, baseline; FPG, BL, baseline; FPG, ofcovariance; analysis 235 / 227 240 / 225 240/225 /235 227 243/232 /235 227 β ; LS, least squares; ; LS,least squares; This articleis protectedbycopyright. Allrightsreserved. Accepted Article

This articleis protectedbycopyright. Allrightsreserved. Accepted Article event. adverse severe SAE, event; AE, adverse orbehaviour. consciousness impairment in to severe due assistance requiring external events were events defined as symptomatic hypoglycaemic mmol/L.measurement Severe <2.9 glucose blood anasymptomatic assistance, or mmol/L noneed forexternal and level <2.9 bloodglucose event with a symptomatic discontinuation AE leadingtreatment topermanent Hypoglycaemia Other significantAE* SAE Treatment-related SAE AE Treatment-related AE Saxagliptin 3. Table (n=231) Physician during the evaluation of safety data. ofsafety evaluation the Physician during the Safety appropriate qualifiedwithGlobalmedically consultation Patient in expert a were identified by AEs *Other significant numberas (%)ofpatients. reported All dataare AE leading todeath Adverse events over 24 Weeks (safety analysis over24Weeks set) events (safety Adverse †

† Hypoglycaemic events were defined as either as were either defined events Hypoglycaemic 78 (32.4) 109 (44.7) 109(44.7) (32.4) 78 04 1(0.4) 1(0.4) 0(0.0) 2(0.8) 50(20.5) 1 (0.4) 3 (1.2) 1 (0.4) 5 (2.1) 8 (3.3) 00 0(0.0) 0 (0.0) 12 4(1.6) 3 (1.2)

Acarbose (n=230) Acarbose

This articleis protectedbycopyright. Allrightsreserved. Accepted Article Figure Figure s 1

This articleis protectedbycopyright. Allrightsreserved. Accepted Article Figure 2

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。 图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具