Effects of Pioglitazone and Metformin Fixed-Dose Combination Therapy on Cardiovascular Risk Markers of Inflammation and Lipid Profile Compared With Pioglitazone and Metformin Monotherapy

O r i g i n a l P a p e r

Effects of Pioglitazone and MetforminFixed-Dose Combination Therapy onCardiovascular Risk Markers of Inflammationand Lipid Profile Compared With Pioglitazoneand Metformin Monotherapy in Patients WithType 2 Diabetes

Alfonso Perez, MD;1 Randal Jacks, MD;2 Vipin Arora, PhD;1

Robert Spanheimer, MD3

Type 2 diabetes mellitus (T2DM) treatmentshould not increase cardiovascular (CV) risk andat best could provide benefit beyond loweringglucose. Pioglitazone has demonstrated afavorable CV profile relative to other oralantidiabetic drugs (OADs) in outcome andobservational studies. This randomized, double-blind, parallel-group controlled study examinedcirculating biomarkers of CV risk in T2DMpatients receiving a fixed-dose combination(FDC) of pioglitazone ⁄ metformin compared withthe respective monotherapies. Patients with stableglycosylated hemoglobin (HbA1c) for 3 monthstaking no OADs were treated with pioglitazone15 mg ⁄ metformin 850 mg FDC twice daily (bid),

pioglitazone 15 mg bid, or metformin 850 mgbid for 24 weeks. FDC and pioglitazoneincreased high-density lipoprotein cholesterol by14.20% and 9.88%, respectively, vs an increaseof 6.09% with metformin (P<.05, metformin vsFDC). Triglycerides decreased with all threetreatments )5.95%, )5.54% and )1.78%,respectively; P=not significant). FDC andpioglitazone significantly decreased smalllow-density lipoprotein and increased largelow-density lipoprotein particle concentrations.Reductions in high-sensitivity C-reactive proteinwere greater in the FDC and pioglitazone groups.Increases in adiponectin were significant in theFDC and pioglitazone groups (P<.0001 vsmetformin). Overall, adverse events were nothigher with the FDC. Thus, treatment with theFDC resulted in improved levels of CVbiomarkers, which were better than or equal tomonotherapy. J Clin Hypertens (Greenwich).2010;12:973–982. ª 2010 Wiley Periodicals, Inc.

Patients with type 2 diabetes mellitus (T2DM)have a significantly greater risk of cardiovascu-

lar (CV) disease (CVD), including coronary heartdisease, stroke, and peripheral vascular disease,1

in addition to the well-known microvascular com-plications such as retinopathy, neuropathy, and

From the Takeda Global Research and DevelopmentCenter, Inc, Deerfield, IL;1 Hill Country MedicalAssociates, New Braunfels, TX;2 and TakedaPharmaceuticals North America, Inc, Deerfield, IL3

Address for correspondence:Alfonso Perez, MD, Takeda Global Research andDevelopment Center, Inc, One Takeda Parkway,Deerfield, IL 60015E-mail: [email protected] received May 20, 2010; revised July 30, 2010;accepted August 20, 2010

doi: 10.1111/j.1751-7176.2010.00389.x

VOL. 12 NO. 12 DECEMBER 2010 THE JOURNAL OF CLINICAL HYPERTENSION 973

nephropathy. While microvascular complicationscontribute to patient morbidity, macrovasculardisease is the leading cause of death (approxi-mately 75%) among patients with type 2 diabe-tes.1–3 Given the association between diabetes andhigh rates of CV events, identification of treatmentapproaches in T2DM that address risk factors forCVD may provide additional benefit for CV healthbeyond the management of blood glucose.

While studies have shown that microvascular com-plications can be reduced by intensive glycemic con-trol in both type 1 and type 2 diabetes,4–6 trials toassess the effect of intensive glycemic control on CVDrisk in diabetes have not yielded conclusive results.7

Although several epidemiologic studies have shown adirect correlation between measures of glycemiccontrol and CV outcomes,8 factors beyond glycemiccontrol on CVD risk must be identified to reducethe potential complications in diabetic patients.

The CV risks associated with T2DM constitutea cluster of comorbidities, including abdominalobesity, insulin resistance, hypertension, and dys-lipidemia. Diabetic dyslipidemia is characterizedby decreased high-density lipoprotein cholesterol(HDL-C), increased triglycerides (TGs), and anincreased number of small dense low-density lipo-protein cholesterol (LDL-C) particles.9,10 The alter-ation in LDL-C particle size and number isassociated with insulin resistance and has beenlinked to an increased risk for CVD.11–14 Patientswith T2DM also have evidence of low-grade sys-temic inflammation, with increased levels of inflam-matory markers such as high-sensitivity C-reactiveprotein (hs-CRP) and decreased levels of the anti-inflammatory molecule adiponectin. This proinflam-matory state is also thought to contribute to theincreased risk of CV complications.15,16 The Fra-mingham Offspring Study showed that elevated hs-CRP levels were related to insulin resistance andthat a patient with the metabolic syndrome and ahigh level of hs-CRP is at increased risk for heartdisease.16 Adiponectin is the most abundant proteinsecreted by adipocytes into the circulation, andreduced adiponectin is associated with the compo-nents of the metabolic syndrome, such as increasedinsulin resistance and blood glucose, decreased fattyacid metabolism, elevated TGs, and reduced HDL-C. Adiponectin also has anti-inflammatory effects,and its gene expression can be activated by peroxi-some proliferator-activated receptor c agonists.17

As previously reported,18 pioglitazone ⁄metfor-min fixed-dose combination (FDC) therapy demon-strated additive efficacy in glucose lowering, withno additive adverse event rate compared with the

respective monotherapies. The objectives of this sec-ond manuscript based on the same study were toevaluate the effects of this FDC therapy on CV riskmarkers, inflammatory markers, and diabetic dysli-pidemia, including lipid fractionations, in type 2diabetes patients.

METHODSPatientsDetailed methods of this study have previouslybeen published.18 Briefly, the trial was a 24-week,prospective, randomized, double-blind, parallel-group study designed to evaluate twice-daily (bid)dosing with the marketed formulation of pioglita-zone 15 mg ⁄metformin 850 mg FDC compared witheach individual component, pioglitazone 15 mgbid and metformin 850 mg bid, in patients withT2DM. All patients provided written informedconsent, and the study was conducted in accor-dance with the Declaration of Helsinki and GoodClinical Practice Guidelines.

Patients who met inclusion criteria were at least18 years of age with a diagnosis of type 2 diabetes,had not received treatment with antidiabetic medi-cation in the 12 weeks prior to screening (otherthan short-term use, ie, �15 days), had a glycosylat-ed hemoglobin (HbA1c) level �7.5% but �10.0%,and had a body mass index �45 kg ⁄m2. Thepatients received counseling on lifestyle modificationfor T2DM including diet and exercise. Exclusioncriteria included New York Heart Association classIII or IV heart failure or history of myocardialinfarction, cerebrovascular accident, percutaneouscoronary intervention, coronary artery bypass graft,or transient ischemic attack in the 6 months priorto screening. Patients received either pioglitazone15 mg ⁄ metformin 850 mg FDC bid, pioglitazone15 mg bid, or metformin 850 mg bid starting onday 1 after randomization and continued to takestudy medication for 24 weeks.

Analytical MethodsAll laboratory samples were collected in accordancewith acceptable laboratory procedures. On days ofstudy visits, patients were required to fast for aminimum of 8 hours prior to all fasting laboratorytests. They were allowed to ingest water and anymedications normally taken at that time of dayexcept study medication until after laboratory sam-ples were obtained. Nothing else by mouth, includ-ing smoking, was allowed.

Lipid panel (total cholesterol, HDL-C, TGs, andLDL-C) and lipid fractionation were measuredat baseline and at the final visit (week 24) or early

THE JOURNAL OF CLINICAL HYPERTENSION VOL. 12 NO. 12 DECEMBER 2010974

termination. Fasting serum lipids, HbA1c values, hs-CRP, and adiponectin were measured centrally atICON Laboratories, Inc (Dublin, Ireland) to ensureconsistent measurement and reduced variability.Direct quantitative enzymatic methods (Roche Diag-nostics, Indianapolis, IN) were used to measure totalcholesterol, HDL-C, TGs, and LDL-C levels in fastingblood samples. Methods used to measure total choles-terol, HDL-C, TGs, and LDL-C were accredited bythe Centers for Disease Control and Prevention’sLipid Standardization Program. Lipoprotein subclassparticle concentrations and sizes were measured usingnuclear magnetic resonance spectroscopy at Lipo-Science (Raleigh, NC). Plasma hs-CRP was measuredby immunoturbidimetry (Roche Diagnostics), andtotal adiponectin was measured by radioimmuno-assay. Laboratory data collected more than 7 daysafter the last dose of study medication were excludedfrom the summaries.

Statistical AnalysisEfficacy variables were analyzed using all randomizedpatients who received at least 1 dose of double-blindstudy medication and who had both a baseline valueand at least 1 value during the treatment period. Allvariable analyses were based on the last-observation-carried-forward (LOCF) method when on-therapydata were missing. The safety analysis set included allpatients who received at least one dose of double-blind study medication.

Treatment group differences in adiponectin, lipidparameters (total cholesterol, HDL-C, TGs, andLDL-C), and lipid fractionations were each ana-lyzed using an analysis of covariance (ANCOVA)model, with treatment group as the factor andbaseline value as a continuous covariate. Signi-ficance tests were based on least squares meansand type III sum-of-squares, using 2-sided tests at a.05 level of significance. All hs-CRP values wereanalyzed using a nonparametric ANCOVA, usingTukey’s normal rank transformation of the percentchange from baseline. The estimate of difference inmedians and 95% confidence intervals was basedon the Hodges–Lehmann method and the distribu-tion-free confidence interval, respectively. Overallsafety and tolerability were assessed with treat-ment-emergent adverse events, clinical laboratorydata, physical examination findings, vital signs, 12-lead electrocardiograms, and weight.

RESULTSPatientsThe majority of patients were white (89%), female(57.7%), and had never smoked (70.3%). There

were no notable differences observed among the 3treatment groups for demographic or baseline char-acteristics, baseline blood pressure, or baseline labo-ratory values for hs-CRP or adiponectin (Table I).Baseline medical conditions and baseline medicationhistory were comparable among the 3 treatmentgroups, with hypertension and dyslipidemia the mostcommon medical conditions at baseline (Table II).

EfficacyThe primary efficacy results from this study havebeen described.18 Briefly, the greatest decrease inHbA1c was observed in the pioglitazone ⁄metforminFDC group, a decrease by 1.83% from a baselineHbA1c of 8.9%. The reduction was statistically sig-nificant compared with those observed with each ofthe monotherapies (P<.0001): reductions from asimilar baseline level by 0.96% and 0.99% for thepioglitazone and metformin monotherapies, respec-tively (P<.0001 compared with the baseline of eachrespective group).

LipidsThe percent increase in HDL-C was largest in thepioglitazone ⁄metformin FDC group (14.20% froma baseline of 41.8 mg ⁄dL; Table III) compared withincreases of 9.88% with pioglitazone (P=.04, piog-litazone vs FDC) and 6.09% with metforminmonotherapy (P<.0001, metformin vs FDC), fromrespective baselines of 43.4 mg ⁄dL and 42.5mg ⁄dL. Moderate decreases from baseline to finalvisit in TGs were seen across all 3 treatmentgroups, with greater decreases in the FDC andpioglitazone groups. However, the differencesbetween the treatment groups were not statisticallysignificant (Table III). The level of LDL-C increasedby 1.19% from baseline in the pioglitazone ⁄metfor-min FDC group and by 6.08% in the pioglitazonegroup (nonsignificant difference), and decreased1.37% in the metformin monotherapy group(P<.05, metformin vs pioglitazone) (Table III).Lipid fractionation showed a decrease in meanLDL-C particle concentration across the 3 treat-ment groups. Mean LDL-C particle size increasedin all groups, but the increases in the pioglitaz-one ⁄metformin FDC and pioglitazone monotherapygroups were statistically significantly greater thanthat in the metformin monotherapy group(0.55 nm and 0.60 nm, respectively, vs 0.20 nm;P<.0001) (Figure 1). Fractionation of LDL-C parti-cle subclasses showed that pioglitazone ⁄metforminFDC and pioglitazone treatment led to a significantdecrease in small-dense LDL particle concentrationvs metformin monotherapy ()319.3 nmol ⁄L and


Table I. Demographic and Baseline Characteristics

Pioglitazone

15 mg ⁄ Metformin

850 mg bid (n=201)

Pioglitazone

15 mg bid (n=189)

Metformin

850 mg bid (n=210)

Total

(N=600)

CharacteristicAge, mean (SD), y 54.7 (12.2) 54.0 (12.1) 53.7 (12.0) 54.1 (12.1)

Sex, No. (%)Female 111 (55.2) 123 (65.1) 112 (53.3) 346 (57.7)

BMI, mean (SD), kg ⁄ m2 30.8 (5.7) 31.2 (5.5) 30.8 (5.7) 30.9 (5.6)Smoking status, No. (%)

Never smoked 132 (65.7) 138 (73.0) 152 (72.4) 422 (70.3)Current smoker 29 (14.4) 24 (12.7) 27 (12.9) 80 (13.3)Former smoker 40 (19.9) 27 (14.3) 31 (14.8) 98 (16.3)

Baseline BPDiastolic BP,mean (SD), mm Hg

n=19877.4 (8.81)

n=17877.6 (9.17)

n=20377.0 (8.73)

Systolic BP,mean (SD), mm Hg

n=198125.9 (14.79)

n=178125.9 (15.18)

n=203125.8 (15.37)

Baseline laboratory valuesa P Valueb

HbA1c, LS mean (SE), % n=201

8.89 (0.07)

n=189

8.69 (0.07)

n=210

8.65 (0.07) .030hs-CRP, median, mg ⁄ L n=198

3.545n=1862.885

n=2083.080 .266

Adiponectin, LSmean (SE), lg ⁄ mL

n=1906.1 (0.34)

n=1756.8 (0.35)

n=1997.2 (0.33) .057

Abbreviations: bid, twice daily; BMI, body mass index; BP, blood pressure; HbA1c, glycosylated hemoglobin; HDL-C,

high-density lipoprotein cholesterol; hs-CRP, high-sensitivity C-reactive protein; LDL-C, low-density lipoprotein cholesterol; LS,least squares; SE, standard error of the mean. aBaseline value was defined as the last observation prior to the first dose ofdouble-blind study medication, including the screen value, if necessary. bP value was from one-way analysis of variance with aterm for treatment.

Table II. Medical Conditions and Concomitant Medications at Baseline and Continuing During the Studya

Pioglitazone

15 mg ⁄ Metformin

850 mg bid (n=201)

Pioglitazone

15 mg bid (n=190)

Metformin

850 mg bid (N=209)b

Baseline medical conditions, No. (%)Hypertension 91 (45.3) 95 (50.0) 94 (45.0)

Hyperlipidemia 90 (44.8) 85 (44.7) 95 (45.5)Heart failure 0 1 (0.5) 1 (0.5)

Concomitant medication category, No. (%)ACEIs ⁄ ARBs 57 (28.4) 69 (36.3) 68 (32.5)

Antiplatelets 33 (16.4) 29 (15.3) 34 (16.3)Lipid-lowering agents 30 (14.9) 26 (13.7) 39 (18.7)Diuretics 27 (13.4) 28 (14.7) 28 (13.4)

NSAIDs 23 (11.4) 23 (12.1) 25 (12.0)b-Blocking agents 19 (9.5) 20 (10.5) 18 (8.6)Calcium channel blockers 17 (8.5) 16 (8.4) 22 (10.5)

Abbreviations: ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; bid, twice daily;NSAIDs, nonsteroidal anti-inflammatory drugs. Hypertension = hypertension and essential hypertension. Hyperlipidemia =hyperlipidemia, dyslipidemia, hypercholesterolemia, hypertriglyceridemia, and lipid metabolism disorder. Heart failure =cardiac failure and cardiac failure congestive. aAll patients who received at least one dose of double-blind study medication

were included in the analyses (safety analysis set). bOne patient randomized to the metformin group received pioglitazonetreatment.


)321.3 nmol ⁄L, respectively, vs )179.0 nmol ⁄L;P<.001) and an increase in concentration of largebuoyant particles (96.0 nmol ⁄L and 115.7 nmol ⁄L

in the pioglitazone ⁄metformin FDC- and pioglita-zone-treated patients, respectively, vs 18.4 nmol ⁄Lwith metformin treatment; P<.0001) (Figure 2).

Table III. Changes from Baseline in Lipid Parameters

Value, LS Means (SE)

Pioglitazone

15 mg ⁄ Metformin

850 mg bid (n=201)

Pioglitazone

15 mg bid (n=189)

Metformin

850 mg bid (N=210)

Total cholesterolBaseline, mg ⁄ dL n=199

195.5 (2.85)

n=186

198.4 (2.95)

n=208

196.0 (2.79)Week 24 ⁄ ET, % change n=178

1.06 (1.183)n=154

4.79 (1.271)an=182

)2.72 (1.169)a,b

HDL-C

Baseline, mg ⁄ dL n=19841.8 (0.76)

n=18443.4 (0.79)

n=20642.5 (0.75)

Week 24 ⁄ ET, % change n=177

14.20 (1.440)

n=152

9.88 (1.552)an=180

6.09 (1.426)c

TriglyceridesBaseline, mg ⁄ dL n=201

202.3 (10.17)

n=189

204.2 (10.49)

n=210

213.7 (9.95)Week 24 ⁄ ET, % change n=180

)5.95 (2.901)n=157

)5.54 (3.107)n=184

)1.78 (2.871)LDL-C

Baseline, mg ⁄ dL n=198118.3 (2.38)

n=184118.9 (2.47)

n=206117.1 (2.34)

Week 24 ⁄ ET, % change n=177

1.19 (2.015)

n=152

6.08 (2.175)

n=180

)1.37 (1.998)d

Abbreviations: bid, twice daily; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol;LS, least squares; SE, standard error of the mean. For early termination (ET), last-observation carried forward was used. aP<.05;bP<.0001 vs pioglitazone monotherapy; cP<.0001 vs pioglitazone ⁄ metformin fixed-dose combination therapy; dP<.05.

Figure 1. Mean changes from baseline in lipid fractionations. Baseline low-density lipoprotein (LDL) particleconcentration for pioglitazone ⁄ metformin fixed-dose combination (FDC), pioglitazone, and metformin were1460.9 nmol ⁄ L, 1517.1 nmol ⁄ L, and 1471.2 nmol ⁄ L, respectively. Baseline LDL particle size were 20.28 nm,20.31 nm, and 20.35 nm, respectively. ***P<.0001 vs pioglitazone ⁄ metformin FDC therapy. ��P<.0001 vspioglitazone monotherapy. SE indicates standard error of the mean.


High-Sensitivity C-Reactive ProteinBaseline median hs-CRP levels were elevated (medi-ans in the 3 groups ranged from 2.9 mg ⁄L to3.5 mg ⁄L) and comparable across treatment groups(Table I; P=.266). Reductions from baseline in hs-CRP were observed as early as week 8 and per-sisted until study end at week 24 ⁄ET, with greatermedian percentage decrease from baseline in thepioglitazone ⁄metformin FDC and pioglitazonemonotherapy groups compared with the metforminmonotherapy group (Figure 3). The reduction frombaseline in hs-CRP for the metformin group wassmaller ()12% to )14%) at week 8 and week 12,except for week 24 ⁄ET ()26.2%), and this wasdue to a larger percentage of LOCF being used toimpute week 24 ⁄ET data for metformin in compar-ison with the pioglitazone ⁄metformin FDC andpioglitazone monotherapy groups.

AdiponectinPioglitazone ⁄metformin FDC and pioglitazonemonotherapy resulted in significant increases frombaseline in adiponectin that were seen at week 16and continued to increase until study end. Theincreases were statistically significant compared with

the minor decreases seen in metformin-treatedpatients at all time points (P<.0001) (Figure 4).

SafetyDetailed safety profiles of the study drugswere described in the primary publication.18 Briefly,pioglitazone ⁄metformin FDC treatment for 24 weekswas well tolerated with no new safety or tolera-bility issues identified for this study population.A total of 312 patients (52.0%) experienced766 adverse events during the course of study.Overall incidences of treatment-emergent adverseevents were similar across all 3 study groups,with 50.7% for pioglitazone ⁄ metformin FDC and52.1% and 53.1% for pioglitazone and metfor-min monotherapies, respectively. There were nodeaths or episodes of congestive heart failure dur-ing the study. Diarrhea was more common inmetformin-treated patients (15.3%) comparedwith 9.0% and 2.6% in the pioglitazone ⁄ metfor-min FDC and pioglitazone groups, respectively.Edema was more common in pioglitazone-treatedpatients (4.2%) vs 3.0% in the pioglitazone ⁄ met-formin FDC patients and 1.4% in metforminmonotherapy patients. Patients had a weight gain

Figure 2. Mean change from baseline in large and small low-density lipoprotein (LDL) particle concentration.Baseline LDL small particle concentrations for pioglitazone ⁄ metformin fixed-dose combination (FDC), pioglitazone,and metformin were 1127.8 nmol ⁄ L, 1163.9 nmol ⁄ L, and 1110.7 nmol ⁄ L, respectively. Baseline LDL large particleconcentrations were 256.5 nmol ⁄ L, 271.1 nmol ⁄ L, and 287.3 nmol ⁄ L, respectively. *P<.001 and **P<.0001 vspioglitazone ⁄ metformin FDC therapy and pioglitazone monotherapy. SE indicates standard error of the mean.


Figure 4. Mean changes from baseline in adiponectin. Baseline adiponectin for pioglitazone ⁄ metformin fixed-dosecombination (FDC), pioglitazone, and metformin were 6.1 lg ⁄ mL, 6.8 lg ⁄ mL, and 7.2 lg ⁄ mL, respectively.*P<.0001 vs pioglitazone ⁄ metformin FDC therapy and pioglitazone monotherapy. SE indicates standard error of themean; ET, early termination.

Figure 3. Median percentage changes from baseline in high-sensitivity C-reactive protein (hs-CRP). Baseline median levelhs-CRP for pioglitazone ⁄ metformin fixed-dose combination (FDC), pioglitazone, and metformin were 3.55 mg ⁄ L,2.89 mg ⁄ L, and 3.08 mg ⁄ L, respectively. �P<.01 vs pioglitazone monotherapy; �P<.001 vs pioglitazone ⁄ metforminFDC therapy; *P<.05 vs pioglitazone ⁄ metformin FDC therapy. ET indicates early termination.


of 1.64 kg on pioglitazone monotherapy and0.69 kg on FDC treatment, while patients onmetformin monotherapy lost 1.28 kg.

DISCUSSIONThe high risk of CVD in patients with T2DMrequires recognition of both traditional as well asnontraditional risk factors that extend beyondhyperglycemia. The current study demonstrated ina population of treatment-naı̈ve patients withT2DM that the combination of pioglitazone andmetformin in an FDC resulted in similar orimproved beneficial effects on circulating biomar-kers of CVD compared with each monotherapy.Pioglitazone ⁄metformin FDC significantly increasedthe HDL-C level compared with each of pioglitaz-one and metformin monotherapy. There was atrend of moderate decrease from baseline in TGlevels with all 3 treatments (nonsignificant differ-ence across three groups). The change from base-line to final visit in LDL-C was a slight increasewith FDC and pioglitazone and a small decrease inthe metformin treatment group. It is important tonote that pioglitazone ⁄metformin FDC and pioglit-azone monotherapy resulted in significant decreasesin small LDL particle concentrations and increasesin large LDL particle concentrations, ie, a signifi-cant shift from small dense particles to large buoy-ant LDL particles. Recent studies have suggestedthat small dense LDL particles may be more ath-erogenic than large LDL particles independent ofLDL-C level.19 This shift in LDL particle number,size, and density, in addition to an increase inHDL-C and decrease in TGs with pioglitazonetreatment, represents a reversal in the pattern ofdiabetic dyslipidemia typically seen in T2DM.

Safety and tolerability were discussed in detail inthe primary publication of the study.18 In brief, theFDC showed safety and tolerability similar to thoseof the two monotherapies at the same doses. Nounexpected safety issues were observed in thisstudy, nor were exacerbations of events that aretypically associated with either monotherapy.

Results from this study were consistent with andextend the post hoc analyses of changes in lipidparameters in the long-term CV outcome Prospec-tive Pioglitazone Clinical Trial in MacrovascularEvents (PROactive) study, which compared pioglit-azone treatment with placebo in subgroups ofpatients on baseline statin and different antihyper-glycemic medications. Pioglitazone treatment led toincreased HDL-C and decreased TG levels indepen-dent of the baseline lipid or antihyperglycemicmedication used.20 Although the 10% reduction in

primary end point in PROactive of all-cause mortal-ity, nonfatal myocardial infarction, stroke, and vari-ous macrovascular interventional procedures didnot reach statistical significance, the 16% decreasein the principal secondary end point of all-causemortality, nonfatal myocardial infarction, and strokedid achieve statistical significance21 (P=.027).

Surrogate markers of atherosclerosis progressionin T2DM patients treated with pioglitazone vsglimepiride have been evaluated in the CarotidIntima-Media Thickness in Atherosclerosis UsingPioglitazone (CHICAGO)22 and Pioglitazone Effecton Regression of Intravascular SonographicCoronary Obstruction Prospective Evaluation(PERISCOPE)23 studies. The CHICAGO studymeasured change in carotid artery intima-mediathickness (CIMT) from baseline as a marker forcoronary atherosclerosis. Results showed that piog-litazone significantly slowed the progression ofmean CIMT changes compared with glimepiride;the increase in HDL-C at 24 weeks predicted thebeneficial effect of pioglitazone on mean CIMT at72 weeks.24 PERISCOPE was conducted in patientswith T2DM and documented coronary arterydisease. Treatment with pioglitazone resulted in asignificantly slower rate of progression of coronaryatherosclerosis (mean change in percent atheromavolume from baseline) compared with glimepiride,as measured by intravascular ultrasound of the cor-onary arteries.23 Similar to the CHICAGO study,the reduction in atheroma volume was correlatedwith changes in HDL ⁄ TG ratios in PERISCOPE.25

In addition to improvements in diabetic dyslipi-demia, changes in circulating biomarkers includedhigher adiponectin and lower hs-CRP levels withpioglitazone ⁄metformin treatment and pioglitazonemonotherapy in the current study. Our resultsshowed that adiponectin, a hormone product ofadipocytes with anti-inflammatory and insulin-sen-sitizing effects, significantly increased from baselinewith pioglitazone ⁄metformin FDC and pioglitazonetreatment compared with metformin monotherapyacross all time points. Several studies have shownthat high plasma adiponectin levels are associatedwith reduced coronary heart disease risk in diabeticpatients, although other studies did not find a sim-ple correlation between adiponectin levels and mac-rovascular disease state.26 The decreases frombaseline in hs-CRP to week 24 ⁄final visit weresignificantly greater with pioglitazone ⁄metforminFDC treatment compared with metformin mono-therapy. Pioglitazone monotherapy also had greaterdecreases than metformin monotherapy; however,the decreases were statistically significant only at


week 8 through week 20. A reduction in hs-CRPlevels was associated with a reduction in CV riskas seen in the Justification for the Use of Statins inPrimary Prevention: An Intervention Trial Evaluat-ing Rosuvastatin trial (JUPITER), which studiedstatin treatment in healthy individuals with elevatedhs-CRP levels at baseline, but without hyperli-pidemia. Statin treatment significantly reduced thelevels of hs-CRP by 37%; it also reduced theincidence of major CV events and death from anycause.27

Concern over potential increased CV event riskswith treatments using rosiglitazone, another mem-ber of the thiazolidinedione drug class, was raisedin a meta-analysis published in 2007.28 It showedthat rosiglitazone treatment was linked to anincreased risk of myocardial infarction and deathfrom CV causes, although similar results have notbeen demonstrated in randomized clinical trials.Multiple observational studies to date across theUnited States, Canada, and Europe have shownthat pioglitazone has a more favorable CV safetyprofile compared with certain other oral antidiabet-ic agents.21,29,30 A recent observational study usinga UK general practice research database examinedcurrently available oral antidiabetic treatments forT2DM on CV safety and found that, comparedwith metformin, sulphonylureas were associatedwith a relatively unfavorable risk profile. Pioglita-zone was associated with a significantly lower riskof all-cause mortality compared with metforminand had a favorable risk profile compared withrosiglitazone.31

Despite clear evidence of benefit for reduction ofmicrovascular disease with glycemic control, to datethere is no such evidence for the role of glycemiccontrol in macrovascular event reductions. Recently,the results of the Action to Control CardiovascularRisk in Diabetes (ACCORD) and Action in Diabe-tes and Vascular Disease: Preterax and Diamicron-MR Controlled Evaluation (ADVANCE) trials sug-gested that intensive glycemic control may notreduce the risk of macrovascular events and mayeven be associated with an increase in risk.32,33

Until the role of glycemic control in macrovasculardisease is understood, management of nonglycemicrisk factors remains equally important in the com-prehensive care of patients with diabetes. Resultsfrom the current study demonstrated an additiveeffect of pioglitazone, on top of metformin and thestandard care for lipid-lowering and antihyperten-sive treatment, on multiple independent CV risk fac-tors: lipids, hs-CRP, and adiponectin. It should beemphasized, however, that improvements in hs-CRP

and adiponectin have not been conclusively demon-strated to be independently associated with areduced risk of macrovascular events. It is alsounclear to what degree the present results can begeneralized to patients with better glycemic controlat baseline or those receiving additional concomi-tant medications.

CONCLUSIONSThe potential benefits of pioglitazone in monothera-py and in combination with metformin on improve-ment of diabetic dyslipidemia appear independentof improved glycemic control. Recognition of CVrisk factors remains important in the care ofpatients with T2DM because of the high incidenceof CVD and the unknown benefit of aggressiveglycemic control in changing CV risk.

Disclosures and acknowledgments: This study is registered asan International Standard Randomized Controlled Trial,number NCT00727857. The study was funded by TakedaGlobal Research and Development Center, Inc, Lake Forest,Illinois. Dr R. Jacks served as investigator on the study andreceived research support. Drs Perez, Spanheimer, and Aroraare employees of Takeda, the study sponsor. The authorswould like to thank Dr Amy Yuping Xia, also an employeeof Takeda, for medical writing, and Manel Valdes-Cruz forediting assistance in the preparation of the manuscript.

REFERENCES

1 Haffner SM, Lehto S, Ronnemaa T, et al. Mortality fromcoronary heart disease in subjects with type 2 diabetesand in nondiabetic subjects with and without priormyocardial infarction. N Engl J Med. 1998;339(4):229–234.

2 Kannel WB, McGee DL. Diabetes and glucose toleranceas risk factors for cardiovascular disease: the Framinghamstudy. Diabetes Care. 1979;2(2):120–126.

3 Stamler J, Vaccaro O, Neaton JD, et al. Diabetes, otherrisk factors, and 12-yr cardiovascular mortality for menscreened in the Multiple Risk Factor Intervention Trial.Diabetes Care. 1993;16(2):434–444.

4 The effect of intensive treatment of diabetes on the devel-opment and progression of long-term complications ininsulin-dependent diabetes mellitus. The Diabetes Controland Complications Trial Research Group. N Engl J Med.1993;329(14):977–986.

5 UK Prospective Diabetes Study Group. Effect of intensiveblood-glucose control with metformin on complications inoverweight patients with type 2 diabetes (UKPDS 34).Lancet. 1998;352(9131):854–865.

6 UK Prospective Diabetes Study Group. Intensive blood-glucose control with sulphonylureas or insulin comparedwith conventional treatment and risk of complications inpatients with type 2 diabetes (UKPDS 33). Lancet. 1998;352(9131):837–853.

7 Skyler JS, Bergenstal R, Bonow RO, et al. Intensive glyce-mic control and the prevention of cardiovascular events:implications of the ACCORD, ADVANCE, and VA Dia-betes Trials: a position statement of the American Diabe-tes Association and a Scientific Statement of the AmericanCollege of Cardiology Foundation and the AmericanHeart Association. J Am Coll Cardiol. 2009;53(3):298–304.


8 Stettler C, Allemann S, Juni P, et al. Glycemic control andmacrovascular disease in types 1 and 2 diabetes mellitus:meta-analysis of randomized trials. Am Heart J. 2006;152(1):27–38.

9 Krauss RM. Lipids and lipoproteins in patients with type2 diabetes. Diabetes Care. 2004;27(6):1496–1504.

10 Siegel RD, Cupples A, Schaefer EJ, et al. Lipoproteins,apolipoproteins, and low-density lipoprotein size amongdiabetics in the Framingham offspring study. Metabolism.1996;45(10):1267–1272.

11 Kuller L, Arnold A, Tracy R, et al. Nuclear magnetic res-onance spectroscopy of lipoproteins and risk of coronaryheart disease in the cardiovascular health study. Arterios-cler Thromb Vasc Biol. 2002;22(7):1175–1180.

12 Mora S, Szklo M, Otvos JD, et al. LDL particle subclass-es, LDL particle size, and carotid atherosclerosis in theMulti-Ethnic Study of Atherosclerosis (MESA). Athero-sclerosis. 2007;192(1):211–217.

13 Gardner CD, Fortmann SP, Krauss RM. Association ofsmall low-density lipoprotein particles with the incidenceof coronary artery disease in men and women. JAMA.1996;276(11):875–881.

14 Stampfer MJ, Krauss RM, Ma J, et al. A prospectivestudy of triglyceride level, low-density lipoprotein particlediameter, and risk of myocardial infarction. JAMA.1996;276(11):882–888.

15 Verma S, Szmitko PE, Ridker PM. C-reactive proteincomes of age. Nat Clin Pract Cardiovasc Med. 2005;2(1):29–36.

16 Rutter MK, Meigs JB, Sullivan LM, et al. C-reactiveprotein, the metabolic syndrome, and prediction of car-diovascular events in the Framingham Offspring Study.Circulation. 2004;110(4):380–385.

17 Yamauchi T, Kadowaki T. Physiological and pathophysio-logical roles of adiponectin and adiponectin receptors inthe integrated regulation of metabolic and cardiovasculardiseases. Int J Obes (Lond). 2008;32(suppl 7):S13–S18.

18 Perez A, Zhao Z, Jacks R, et al. Efficacy and safety ofpioglitazone ⁄ metformin fixed-dose combination therapycompared with pioglitazone and metformin monotherapyin treating patients with T2DM. Curr Med Res Opin.2009;25(12):2915–2923.

19 Rizzo M, Berneis K. Small, dense low-density-lipoproteinsand the metabolic syndrome. Diabetes Metab Res Rev.2007;23(1):14–20.

20 Spanheimer R, Betteridge DJ, Tan MH, et al. Long-termlipid effects of pioglitazone by baseline anti-hyperglycemiamedication therapy and statin use from the PROactiveexperience (PROactive 14). Am J Cardiol. 2009;104(2):234–239.

21 Dormandy JA, Charbonnel B, Eckland DJ, et al. Second-ary prevention of macrovascular events in patients with

type 2 diabetes in the PROactive Study (PROspectivepioglitAzone Clinical Trial In macroVascular Events): arandomised controlled trial. Lancet. 2005;366(9493):1279–1289.

22 Mazzone T, Meyer PM, Feinstein SB, et al. Effect of piog-litazone compared with glimepiride on carotid intima-media thickness in type 2 diabetes: a randomized trial.JAMA. 2006;296(21):2572–2581.

23 Nissen SE, Nicholls SJ, Wolski K, et al. Comparison ofpioglitazone vs glimepiride on progression of coronaryatherosclerosis in patients with type 2 diabetes: thePERISCOPE randomized controlled trial. JAMA. 2008;299(13):1561–1573.

24 Davidson M, Meyer PM, Haffner S, et al. Increasedhigh-density lipoprotein cholesterol predicts the pioglita-zone-mediated reduction of carotid intima-media thick-ness progression in patients with type 2 diabetes mellitus.Circulation. 2008;117(16):2123–2130.

25 Nicholls SJ, Bayturan O, Hu T, et al. Abstract 5052: low-ering of the triglyceride ⁄ HDL cholesterol ratio predictsthe benefit of pioglitazone on progression of coronary ath-erosclerosis in diabetic patients. Circulation. 2010;118:S1135.

26 Antoniades C, Antonopoulos AS, Tousoulis D, et al.Adiponectin: from obesity to cardiovascular disease. ObesRev. 2009;10(3):269–279.

27 Ridker PM, Danielson E, Fonseca FA, et al. Rosuvastatinto prevent vascular events in men and women withelevated C-reactive protein. N Engl J Med. 2008;359(21):2195–2207.

28 Nissen SE, Wolski K. Effect of rosiglitazone on the risk ofmyocardial infarction and death from cardiovascularcauses. N Engl J Med. 2007;356(24):2457–2471.

29 Lincoff AM, Wolski K, Nicholls SJ, et al. Pioglitazoneand risk of cardiovascular events in patients with type 2diabetes mellitus: a meta-analysis of randomized trials.JAMA. 2007;298(10):1180–1188.

30 Rohatgi A, McGuire DK. Effects of the thiazolidinedionemedications on micro- and macrovascular complicationsin patients with diabetes–update 2008. Cardiovasc DrugsTher. 2008;22(3):233–240.

31 Tzoulaki I, Molokhia M, Curcin V, et al. Risk of cardio-vascular disease and all cause mortality among patientswith type 2 diabetes prescribed oral antidiabetes drugs:retrospective cohort study using UK general practiceresearch database. BMJ. 2009;339:b4731.

32 Gerstein HC, Miller ME, Byington RP, et al. Effects ofintensive glucose lowering in type 2 diabetes. N Engl JMed. 2008;358(24):2545–2559.

33 Patel A, MacMahon S, Chalmers J, et al. Intensive bloodglucose control and vascular outcomes in patients withtype 2 diabetes. N Engl J Med. 2008;358(24):2560–2572.


Documents

Effects of Pioglitazone and Metformin Fixed-Dose Combination Therapy on Cardiovascular Risk Markers of Inflammation and Lipid Profile Compared With Pioglitazone and Metformin Monotherapy