Intensive Treat-to-Target StatinTherapy in High-Risk JapanesePatients WithHypercholesterolemia andDiabetic Retinopathy: Report ofa Randomized StudyDiabetes Care 2018;41:1275–1284 | https://doi.org/10.2337/dc17-2224

OBJECTIVE

Diabetes is associated with high risk of cardiovascular (CV) events, particularly inpatients with dyslipidemia and diabetic complications. We investigated the in-cidence of CV events with intensive or standard lipid-lowering therapy in patientswith hypercholesterolemia, diabetic retinopathy, and no history of coronary arterydisease (treat-to-target approach).

RESEARCH DESIGN AND METHODS

In this multicenter, prospective, randomized, open-label, blinded end point study,eligible patients were randomly assigned (1:1) to intensive statin therapy targetingLDL cholesterol (LDL-C) <70 mg/dL (n = 2,518) or standard statin therapy targetingLDL-C 100–120 mg/dL (n = 2,524).

RESULTS

Mean follow-up was 376 13 months. LDL-C at 36months was 76.56 21.6 mg/dL inthe intensive group and 104.16 22.1 mg/dL in the standard group (P < 0.001). Theprimary end point events occurred in 129 intensive group patients and 153 standardgroup patients (hazard ratio [HR] 0.84 [95% CI 0.67–1.07]; P = 0.15). The relationshipbetween the LDL-C difference in the two groups and the event reduction rate wasconsistent with primary prevention studies in patients with diabetes. Exploratoryfindings showed significantly fewer cerebral events in the intensive group (HR 0.52[95% CI 0.31–0.88]; P = 0.01). Safety did not differ significantly between the twogroups.

CONCLUSIONS

We found no significant decrease in CV events or CV-associated deaths with intensivetherapy, possibly because our between-group difference of LDL-C was lower thanexpected (27.7 mg/dL at 36 months of treatment). The potential benefit of achievingLDL-C <70 mg/dL in a treat-to-target strategy in high-risk patients deserves furtherinvestigation.

1Department of Endocrinology, Metabolism andNephrology, Keio University School of Medicine,Tokyo, Japan2Department of Cardiovascular Medicine, TheUniversity of Tokyo Graduate School of Medi-cine, Tokyo, Japan3Department of Clinical Medicine (Biostatisticsand Pharmaceutical Medicine), School of Phar-macy, Kitasato University, Tokyo, Japan4Department of Cardiovascular Medicine, Wa-kayama Medical University, Wakayama, Japan5Department of Cardiovascular Medicine, Grad-uate School of Medicine Juntendo University,Tokyo, Japan6Sakura Hospital, Fukuoka, Japan7Department of Cardiology, Saitama MedicalCenter, Jichi Medical University, Saitama, Japan8Department of Integrated Medicine and Informat-ics, Ehime University Graduate School of Medicine,Toon, Japan9Division of CardiovascularMedicine, Departmentof Internal Medicine, Kobe University GraduateSchool of Medicine, Kobe, Japan10Division of Endocrinology and Metabolism,Department of Internal Medicine, Jichi MedicalUniversity, Shimotsuke, Japan11Division of Cardiology, Cardiovascular Center,Ageo Central General Hospital, Ageo, Japan12Division of Nephrology, Endocrinology and Vas-cular Medicine, Tohoku University Graduate Schoolof Medicine, Sendai, Japan

Hiroshi Itoh,1 Issei Komuro,2

Masahiro Takeuchi,3 Takashi Akasaka,4

Hiroyuki Daida,5 Yoshiki Egashira,6

Hideo Fujita,7 Jitsuo Higaki,8 Ken-ichi Hirata,9

Shun Ishibashi,10 Takaaki Isshiki,11

Sadayoshi Ito,12 Atsunori Kashiwagi,13

Satoshi Kato,14 Kazuo Kitagawa,15

Masafumi Kitakaze,16 Takanari Kitazono,17

Masahiko Kurabayashi,18

Katsumi Miyauchi,19 Tomoaki Murakami,20

Toyoaki Murohara,21 Koichi Node,22

Susumu Ogawa,23 Yoshihiko Saito,24

Yoshihiko Seino,25 Takashi Shigeeda,26

Shunya Shindo,27 Masahiro Sugawara,28

Seigo Sugiyama,29 Yasuo Terauchi,30

Hiroyuki Tsutsui,31 Kenji Ueshima,32

Kazunori Utsunomiya,33

Masakazu Yamagishi,34

Tsutomu Yamazaki,35 Shoei Yo,36

Koutaro Yokote,37 Kiyoshi Yoshida,38

Michihiro Yoshimura,39

Nagahisa Yoshimura,40 KazuwaNakao,41 and

Ryozo Nagai,42 for the EMPATHY

Investigators

Diabetes Care Volume 41, June 2018 1275

CARDIOVASCULA

RANDMETA

BOLIC

RISK

Dyslipidemia and impaired glucose me-tabolism in diseases such as diabetes areknown risk factors for cardiovascular dis-ease; patients with both conditions are ateven greater risk of cardiovascular (CV)events (1–3). Meta-analysis (4) has asso-ciated lower levels of LDL cholesterol(LDL-C) with reduced risk of CV events inpatients with type 2 diabetes, and the riskof CV events increases further in patientswith diabetic retinopathy (5,6). However,little evidence is currently available onthe efficacy of lipid therapy specifically inthese very high-risk patients.The association of lower LDL-C and

reduced risk of CV events has spurredinterest in a treat-to-target approach,adjusting the drug dose to achieve aspecific LDL-C target. However, at thetime of this study, almost all large-scalelipid-lowering clinical studies with statinswere either placebo controlled or com-pared treatment results based on statintype or dose (7). The authors of the 2013American College of Cardiology (ACC)/American Heart Association (AHA) guide-line on lipid management (8) thus con-cluded that evidence was insufficient toprove the benefits of treat-to-target anddid not include specific treat-to-targetlevels.To explore the potential benefits of

treat-to-target in very high-risk patients,

we conducted a large-scale clinical studyin patients with hypercholesterolemia anddiabetic retinopathy (type 2 diabetes). Fewprospective clinical studies have evaluatedthe efficacy of intensive lipid-lowering ther-apy specifically in primary prevention pa-tientswith diabetes, particularly for reducingrisk through lipid-lowering intervention inhypercholesterolemic patients with dia-betic retinopathy.

The standard versus intEnsive statintherapy for hypercholesteroleMic Patientswith diAbetic retinopaTHY (EMPATHY)study examined whether intensive lipid-lowering therapy is superior to standardtherapy in reducing the incidence of CVevents in patients with hyperlipidemiaand diabetic retinopathy but no historyof coronary artery disease. Patients weredivided into two groups targeting differ-ent LDL-C levels (,70 or $100 and,120 mg/dL). Statin therapy, regardlessof statin type, was used to control LDL-Cat the targeted level in each group. Safetyand efficacy were compared betweengroups.

The EMPATHY study is the first to assessthe benefits of intensive versus standardstatin therapy for patients with hypercho-lesterolemia and diabetic retinopathy in aprimary prevention setting. The study alsoevaluates the appropriateness of treat-to-target, because all patients were treated

to achieve specific LDL-C targets by titrat-ing statin therapy.

RESEARCH DESIGN AND METHODS

Study DesignThe study used a multicenter, prospec-tive, randomized, open-label, blinded endpoint (PROBE) design (9) and enrolled pa-tients at hospitals and family practice clin-ics across Japan. The study was conductedunder the Declaration of Helsinki and Jap-anese ethical guidelines for clinical studies.The protocol was reviewed and approvedby the institutional review board of eachparticipating center.

PatientsPatients who had elevated LDL-C anddiabetic retinopathy without a historyof coronary artery disease were eligiblefor participation (Supplementary Data).All patients were enrolled by the inves-tigators and provided written informedconsent.

Randomization and MaskingA data center provided the computer-generated allocation sequence, stratifiedby sex, age, and baseline hemoglobin A1c(HbA1c). After patient eligibility was con-firmed, the investigator contacted the datacenter for the allocated treatment. Staff whogenerated the allocation sequence were notinvolved in patient enrollment.

13Kusatsu General Hospital, Kusatsu, Japan14Department of Ophthalmology, The Universityof Tokyo Graduate School of Medicine, Tokyo,Japan15Department of Neurology, Tokyo Women’sMedical University School of Medicine, Tokyo,Japan16Division of Cardiology, National Cerebral andCardiovascular Center, Suita, Japan17Department of Medicine and Clinical Science,Graduate School of Medical Sciences, KyushuUniversity, Fukuoka, Japan18Department of Medicine and Biological Science,Gunma University Graduate School of Medicine,Maebashi, Japan19Department of Cardiology, Graduate School ofMedicine Juntendo University, Tokyo, Japan20Department of Ophthalmology, Kyoto Univer-sity Graduate School of Medicine, Kyoto, Japan21Department of Cardiology, Nagoya UniversityGraduate School of Medicine, Nagoya, Japan22Department of Cardiovascular Medicine, SagaUniversity, Saga, Japan23Division of Nephrology, Endocrinology and Vas-cular Medicine, Tohoku University Hospital, Sendai,Japan24First Department of Internal Medicine, NaraMedical University, Kashihara, Japan25Department of Cardiology, NipponMedical SchoolChiba Hokuso Hospital, Inzai, Japan

26Ideta Eye Clinic, Kumamoto, Japan27Department of Cardiovascular Surgery, TokyoMedical University Hachioji Medical Center,Hachioji, Japan28Sugawara Medical Clinic, Tokyo, Japan29Department of Cardiology, Jinnouchi Hospital,Kumamoto, Japan30Department of Endocrinology and Metabo-lism, Yokohama City University School of Med-icine, Yokohama, Japan31Department of Cardiovascular Medicine, Facultyof Medical Sciences, Kyushu University, Fukuoka,Japan32Department of EBM Research, Institute for Ad-vancement of Clinical and Translational Science,Kyoto University Hospital, Kyoto, Japan33Division of Diabetes, Metabolism and Endo-crinology, Department of Internal Medicine, TheJikei University School of Medicine, Tokyo, Japan34Department of Cardiovascular and Internal Med-icine, Kanazawa University Graduate School ofMedicine, Kanazawa, Japan35Clinical Research Support Center, The Univer-sity of Tokyo Hospital, Tokyo, Japan36Yo Clinic, Kyoto, Japan37Department of Clinical Cell Biology and Med-icine, Chiba University Graduate School of Med-icine, Chiba, Japan38SakakibaraHeart Institute ofOkayama,Okayama,Japan

39Division of Cardiology, Department of InternalMedicine, The Jikei University School of Medi-cine, Tokyo, Japan40Kitano Hospital, The Tazuke Kofukai Medical Re-search Institute, Osaka, Japan41Medical Innovation Center, KyotoUniversity Grad-uate School of Medicine, Kyoto, Japan42Jichi Medical University, Shimotsuke, Japan

Corresponding author: Hiroshi Itoh, hiito@keio.jp.

Received 24 October 2017 and accepted 12March 2018.

Clinical trial reg. no. UMIN000003486, www.umin.ac.jp/ctr/.

This article contains Supplementary Data onlineat http://care.diabetesjournals.org/lookup/suppl/doi:10.2337/dc17-2224/-/DC1.

© 2018 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.

1276 Intensive Statin Therapy in High-Risk Patients Diabetes Care Volume 41, June 2018

This study was designed to evaluatethe treat-to-target approach, so investiga-tors could not be blinded to statin typesand doses. We thus selected a PROBEdesign rather than a double-blind design.To eliminate bias, defined end pointswere evaluated by a blinded end pointcommittee.

ProceduresAs previously described (9), this PROBEstudy constituted a run-in period (4–8weeks) and a treatment period (2–5.5years). During run-in, patients receivedoral statin monotherapy targeting LDL-Cbetween 100 and 120 mg/dL. Patientswere then assessed for eligibility andrandomly assigned (1:1) to oral intensivetherapy (targeting LDL-C ,70 mg/dL) orstandard therapy (targeting LDL-C be-tween 100 and 120 mg/dL). LDL-C levelswere calculated from the Friedewald for-mula. LDL-C targets were based on guide-lines in the U.S. and Japan when thestudy was designed (10,11). Each inves-tigator independently selected a statinfor the run-in and treatment periodsfor each patient. Statin dose escalationand switching to another statin werepermitted in both groups. The investi-gators generally measured LDL-C oncemonthly for 6 months after study start toconfirm that the treatment target hadbeen reached. After that point, LDL-Cwas measured every 6 months if valuesremained within the treatment target, ormore frequently at the discretion of thephysician (for example, after LDL-C ex-ceeded the target treatment range, toconfirm that the treatment target hadbeen reacquired) (9).Concomitant lipid-lowering therapy was

prohibited with fibrates, ezetimibe, ethylicosapentate, anion exchange resins, pro-bucol, nicotinic acid derivatives, phytoster-ols, elastase, dextran sulfate sodium sulfur,pantethine, or polyenephosphatidylcholine.Details of treatment for diabetes and

hypertension were provided previously,along with medical histories, baselinefindings, and changes during the treat-ment period in parameters such as bodyweight, blood pressure, blood chemistry,and electrocardiogram (9). Statin treat-ment adherence, concomitant use of otherdrugs, and adverse events (AEs) were in-vestigated periodically throughout thestudy. The physician determined adher-ence by questioning each patient duringeach visit to the hospital and recording

the patient’s answers. No situationsarose during the study that requiredconsideration of study discontinuationor suspension.

OutcomesThe primary and secondary outcomesof the EMPATHY study were previouslydescribed (9) (Supplementary Data). Theprimary outcome was the composite in-cidence of CV events, including cardiac,cerebral, renal, and vascular events, orCV-associated death. The secondary out-comes included death from any cause,individual incidence of study-defined CVevents for the primary end point, inci-dence of stroke, change in laboratoryvariables related to chronic kidney dis-ease (CKD), and safety. Items assessed forsafety are listed in the SupplementaryData. Primary and secondary end pointswere adjudicated by an event evaluationcommittee whose members were un-aware of the treatment allocation.

In the previous century, coronary arterydisease and stroke were the primary endpoints in most large-scale clinical studies ofstatins. However, the relationship betweenCKD and arteriosclerosis began to be morewidely accepted from about 2000. Todayischemic heart disease, cerebrovasculardisease, peripheral vascular disease, andrenal impairment are widely understoodto be ischemic conditions, and a meta-analysis of randomized control and cross-over studies has shown that statins inhibitproteinuria and progression of nephropa-thy (12). We thus selected a range ofprimary end points based on arterioscle-rosis, including renal events.

Statistical AnalysisStatistical methods for the EMPATHY studywere previously described (9). The plannedsample size was 5,000 patients, with2,500 in each treatment group. Thiswas considered sufficient to detect a haz-ard ratio (HR) of 0.65 for the superiority ofintensive therapy with a power of 80%and a two-sided significance level of 5%.The incidence of CV events during the3-year treatment period was estimatedat 2.7% for intensive therapy and 4.1%for standard therapy, based on earlierreports in the literature (13–16). Weusedtheseestimates tocalculate thesam-ple size, assuming a withdrawal rate of15% and a study period of 4.5 years. Weestimated 179 occurrences of the pri-mary end point by the end of the study.

The full analysis set (FAS) was selectedas the main analysis set for efficacy. Theprimary and secondary end points wereanalyzed in the same way in the FAS andthe per protocol set, and results for thetwo sets were examined for consistency.The FAS comprised all randomly allocatedsubjects for whom efficacy data wereavailable. The per protocol set comprisedall members of the FAS, except for sub-jects who did not qualify or were nottreated and cases of protocol violation ornoncompliance. All subjects who receivedthe study treatment at least once and forwhom safety information was availablewere included in the safety analysis set.In all analyses, each subject was includedin the allocation group.

We compared the primary end pointbetween the treatment groups by log-ranktest, stratified by sex, age, and baselineHbA1c, and used a stratified Cox propor-tional hazards model to estimate the HRand 95% CI.

Interim analysis was performed duringthe study at a prespecified time point,adjusted by the Lan-DeMets a spendingfunctionwithO’Brien-Fleming boundaries.An independent data monitoring commit-tee assessed the analysis results.

The software used was SAS version 9.2(SAS Institute).

RESULTS

Study PatientsA total of 5,995 patients were enrolledbetween May 2010 and October 2013(Supplementary Fig. 1) at 772 primaryprevention sites (323 hospitals and449 clinics) in Japan. Of these patients,5,144 were randomized to intensive(2,571) or standard (2,573) statin ther-apy; 53 in intensive therapy and 49 instandard therapy were subsequently ex-cluded from data analysis. Data were an-alyzed from 5,042 patients (2,518 in theintensive therapy group and 2,524 in thestandard therapy group). Mean follow-upwas 37 6 13 months.

Baseline CharacteristicsPatients randomized to the two treatmentgroups had similar baseline characteristics(Table 1). Atorvastatin, rosuvastatin, andpitavastatin users accounted for 54.8% ofall patients in the intensive group and54.4% in the standard group at baseline.The other patients used pravastatin, flu-vastatin, or simvastatin. At the end of thestudy, the proportion of patients using

care.diabetesjournals.org Itoh and Associates 1277

atorvastatin, rosuvastatin, and pitavasta-tin increased (95.6%) in the intensivegroup but remained nearly unchanged(53.4%) in the standard group. For allstatin types, dose at baseline was nearlyidentical between the intensive and stan-dard groups. By the last visit, statin dosehad increased for all statin types in the

intensive group; the mean doses for thestandard group remained essentially un-changed over the course of the study(Supplementary Table 1). It should benoted that the statin dose for “intensive”therapy in Japan is lower than in theU.S. and Europe; at the last visit in thisstudy, the dose in the intensive group

would be considered “moderate to low-intensity” statin therapy under ACC/AHAguidelines.

Laboratory ValuesThe mean level of LDL-C for the patientsin the intensive group was 106.2 626.7 mg/dL (median 105.0 mg/dL) at thestart of treatment. That level decreasedto 85.6 6 24.3 mg/dL (83.0 mg/dL) after6 months of treatment and continuedto gradually decline over time, reaching76.5 6 21.6 mg/dL (73.0 mg/dL) at36 months of treatment. For patientstreated with standard statin therapy,mean LDL-C was 106.1 6 25.9 mg/dL(105.0 mg/dL) at the start of treat-ment and remained at or near that levelthroughout the treatment period. Thedifference between the two groups was23.6 mg/dL at 1 year and 27.7 mg/dL at3 years. The difference was significantfrom 6 to 60 months after the start oftreatment (Supplementary Fig. 2).

Changes in other lipid parameters areshown in Supplementary Table 2. Totalcholesterol and triglycerides were lowerin the intensive than the standard group.HDL-C values differed very little betweenthe two groups.

Blood pressure, HbA1c, serum creati-nine, creatinekinase, andhs-CRPare sum-marized in Supplementary Fig. 3. Overallfindings for each treatment group dif-feredlittleovertime, includingthesepara-meters or hematological results, or liverand renal function test results.After treat-ment for 1 year and thereafter, mean hs-CRPwas significantly lower in the intensivegroup than the standard group.

Efficacy End PointsThe primary outcome, combined incidenceof CV events or deaths associated with CVevents, was lower in the intensive group(LDL-C target ,70 mg/dL, 129 patients)than the standard group (LDL-C target$100 and ,120 mg/dL, 153 patients),but that difference was not statisticallysignificant (HR 0.84 [95% CI 0.67–1.07];P = 0.15) (Fig. 1). Calculated values for theprimary and secondary outcomes are listedin Supplementary Table 3.

The incidence of deaths from anycause did not differ significantly betweenthe two groups (HR 1.21 [95% CI 0.77–1.91]) (Fig. 2). The incidence of cerebralevents (cerebral infarction or cerebralrevascularization) decreased significantlyin the intensive group (HR 0.52 [95% CI

Table 1—Demographic characteristics (FAS)

CharacteristicsIntensive therapy

(n = 2,518)Standard therapy

(n = 2,524)

Male 1,200 (47.7) 1,203 (47.7)

Age (years)* 63.0 6 10.8 63.2 6 10.4

Height (cm) 158.9 6 9.6 159.0 6 9.5

Weight (kg) 65.2 6 13.9 64.9 6 13.7

BMI† 25.7 6 4.3 25.6 6 4.4

Abdominal circumference (cm) 90.4 6 10.6 90.4 6 11.1

Lipid-lowering agents‡None 1,105 (43.9) 1,040 (41.2)1 drug 1,406 (55.8) 1,481 (58.7)$2 drugs 7 (0.3) 3 (0.1)

Statin‡No 1,201 (47.7) 1,155 (45.8)Yes 1,317 (52.3) 1,369 (54.2)

Smoking§ 462 (18.3) 480 (19.0)

Family historyDiabetes 1,334 (53.0) 1,308 (51.8)Coronary artery disease 325 (12.9) 318 (12.6)Cerebrovascular disease 494 (19.6) 530 (21.0)

Duration of diabetes (years) 12.8 6 8.6 13.0 6 9.0

Diabetic complicationsNeuropathy 777 (30.9) 781 (30.9)Nephropathy 1,358 (53.9) 1,290 (51.1)

Hypertension 1,777 (70.6) 1,797 (71.2)

Peripheral artery disease (Fontaine class I) 125 (5.0) 110 (4.4)

Other complications and medical history 1,971 (78.3) 2,021 (80.1)

Compliance with statin use from provisionalenrollment to full enrollment

None 13 (0.5) 5 (0.2),50 11 (0.4) 6 (0.2)50–75 36 (1.4) 28 (1.1)$75 2,448 (97.2) 2,477 (98.1)

Funduscopy|Simple retinopathy 1,683 (66.8) 1,669 (66.1)Preproliferative retinopathy 423 (16.8) 485 (19.2)Proliferative retinopathy 390 (15.5) 355 (14.1)Other¶ 16 (0.6) 10 (0.4)

HbA1c (%)* 7.8 6 1.3 7.8 6 1.3

LDL-C (mg/dL)# 106.2 6 26.7 106.1 6 25.9

Blood pressure (mmHg)Systolic 134.6 6 16.8 134.6 6 16.3Diastolic 74.9 6 11.5 74.8 6 11.1

Estimated glomerular filtration rate(mL/min/1.73 m2) 74.0 6 20.6 74.6 6 20.3

Data are mean6 SD or n (%). *Values were obtained at the time of consent. †BMI is the weight inkilograms divided by the square of the height in meters. ‡Values were obtained at provisionalenrollment. §Not including past smokers. |Diagnosed by ophthalmologists based on themodified Davis classification. ¶Includes 17 patients who had a history of laser therapy but nofunduscopic findings at enrollment. The remaining nine patients were found to be retinopathynegative after enrollment. #Values were calculated using the Friedewald equation; LDL-C = totalcholesterol 2 (HDL-C 1 triglyceride/5).

1278 Intensive Statin Therapy in High-Risk Patients Diabetes Care Volume 41, June 2018

0.31–0.88]) (Fig. 1), as did the incidenceof cerebral infarction (HR 0.54 [95%CI 0.32–0.90]) (Fig. 2). The incidence ofstroke (cerebral infarction, cerebral hem-orrhage, and subarachnoid hemorrhage),cerebral hemorrhage, and subarachnoidhemorrhage did not differ between thegroups (HR 0.64 [95% CI 0.40–1.01], HR1.34 [95% CI 0.46–3.86], and HR NA [notapplicable], respectively). No other sig-nificant differences were noted. The HRfor cardiac events was 0.93 (95% CI 0.65–1.33), for renal events 1.07 (95% CI 0.72–1.58), and for vascular events 1.00 (95%CI 0.38–2.67) (Supplementary Table 3).Change and percent change from

baseline to 36 months in the parametersfor CKD were evaluated as a secondaryend point. Estimated glomerular filtrationrate decreased by 5.9 mL/min/1.73 m2

(7.8%) in the intensive group and 6.5mL/min/1.73 m2 (8.3%) in the standardgroup.Urine albumin increasedby 85.3 mg/g

creatinine (343.2%) and 86.5 mg/g creat-inine (300.8%), and urine protein in-creased 10.7 mg/dL (1,503.1%) and11.0 mg/dL (236.1%) in the intensiveand standard groups, respectively. Noneof these parameters differed significantlybetween the intensive and standard groups(Supplementary Table 4).

As part of an exploratory analysis, weperformed subgroup analysis of the pri-mary end point for both treatmentgroups based on various demographicfactors (Supplementary Fig. 4). No sig-nificant heterogeneity of treatment ef-fects was seen in any of the factorsexamined.

SafetyOverall, a similar percentage of patientsin both groups experienced AEs (inten-sive 75.3%; standard 75.2%) and seriousAEs (intensive 21.3%; standard 22.0%)(Table 2). A significantly higher proportion

of patients experienced adverse drug re-actions (ADRs) in the intensive group(10.1%) than the standard group (6.7%)(P, 0.001), butmost of those ADRsweremild. The proportion of serious ADRs wassimilar between groups (1.3% vs. 0.9%;P = 0.22).

CONCLUSIONS

The EMPATHY study assessed the bene-fits of intensive statin monotherapy forlipid management in patients with hyper-cholesterolemia and diabetic retinopathyin a primary prevention setting. The studyalso evaluated the appropriateness of thetreat-to-target approach in this patientpopulation. Results indicated that lipid-lowering therapy targeting LDL-C,70mg/dLhadnomorebeneficial effect on theprimaryend point than therapy targeting LDL-C of100–120mg/dL. In exploratoryfindings, thesecondary end points of cerebral events and

Figure 1—Cumulative event curve for the primary end point (A) and HR of the primary and secondary end points (each component of primary end points)in the intensive and standard therapy groups (B). Cardiac events includedmyocardial infarction, unstable angina requiring unscheduled hospitalization, orcoronary revascularization (percutaneous coronary intervention or coronary artery bypass grafting). Cerebral events included cerebral infarction orcerebral revascularization. Renal events included initiation of chronic dialysis or increase in serum creatinine level by at least twofold (and.1.5 mg/dL).Vascular events included aortic disease or peripheral arterial disease (aortic dissection, mesenteric artery thrombosis, severe lower-limb ischemia[ulceration], revascularization, or finger/lower-limb amputation caused by arteriosclerosis obliterans). P value was calculated using a stratified log-ranktest with sex (male or female), age (,60 or$60 years), and baseline HbA1c (,8.4 or$8.4 [NGSP%]) as stratification factors. HR (95% CI) was estimatedusing a stratified Cox proportional hazards model with sex (male or female), age (,60 or $60 years), and baseline HbA1c (,8.4 or $8.4 [NGSP%]) asstratification factors; P values for secondary end points are nominal.

care.diabetesjournals.org Itoh and Associates 1279

cerebral infarction were reduced significantlyin the intensive group. No major safetyconcerns were noted.The Improved Reduction of Out-

comes: Vytorin Efficacy International Trial

(IMPROVE-IT) (17) demonstrated the use-fulness of intensive lipid therapy for thesecondary prevention of CV events, usingstatin and ezetimibe to achieve LDL-Cas low as 50 mg/dL. Our study focused

on primary prevention to expand theevidence regarding intensive lipid man-agement using statinmonotherapy. How-ever, our results differed from thoseprevious studies using statins.

Figure 2—Cumulative event curve for secondary end points in the intensive and standard therapy groups: death from any cause (A), stroke (B), cerebralinfarction (C), cerebral hemorrhage (D), and subarachnoid hemorrhage (E). P value was calculated using a stratified log-rank test with sex (male orfemale), age (,60 or$60 years), and baseline HbA1c (,8.4 or$8.4 [NGSP %]) as stratification factors. HR (95% CI) was estimated using a stratified Coxproportional hazards model with sex (male or female), age (,60 or $60 years), and baseline HbA1c (,8.4 or $8.4 [NGSP %]) as stratification factors;P values for secondary end points are nominal.

1280 Intensive Statin Therapy in High-Risk Patients Diabetes Care Volume 41, June 2018

Why did our study fail to show superiorresults for intensive therapy? EMPATHYincluded some soft end points that re-quired subjective assessment, such as re-vascularization, and renal events thatwere attributable to arteriosclerosis. Toevaluate these effects on the study results,we compared the groups using three-point major CV events (CV death, nonfatalmyocardial infarction, and nonfatal stroke)and the primary end point after excludingrenal events. We found no significantdifferences (HR 0.82 [95% CI 0.58–1.14]and HR 0.76 [95% CI 0.57–1.01], respec-tively) (Supplementary Fig. 5), suggestingthat our study findings were unaffectedby either soft end points or renal events.Primary end point incidence was 20.41/

1,000 person-years in the standard groupand 17.27/1,000 person-years in the in-tensive group. The cumulative incidenceat 3 years was 5.8% and 5.4%, respec-tively, exceeding our earlier estimates of4.1% and 2.7%. The statistical power ofthe study was thus sufficient to detectsignificant differences in CV events be-tween the two groups.We believe that this study failed to

find a significant reduction in the primaryend point because of the smaller-than-predicted difference in LDL-C betweenthe two treatment groups. Our planned

between-group difference in LDL-C was;40mg/dL (,70mg/dL for the intensivegroup vs. ;110 mg/dL for the standardgroup), and the HR was predicted to be0.65. However, after 3 years of treatment,the actual LDL-C difference was 27.7mg/dL(76.5 vs. 104.1 mg/dL).

In exploratory findings, we comparedour results to the Cholesterol TreatmentTrialists’ (CTT) meta-analysis (18) andthree primary prevention studies in pa-tients with diabetes (13–15). We graphedthe relationship between changes inLDL-C and proportional reduction in eventrate (major CV events including majorcoronary events [coronary death or non-fatal myocardial infarction], stroke ofany type, and coronary revascularization[angioplasty or bypass grafting]). Inter-estingly, the slope of the line for the di-abetes studies seems to be steeper thanthe original CTT line (Supplementary Fig.6). Next, we plotted the findings of theEMPATHY study. In our study, the predic-ted between-group difference in LDL-C of;40 mg/dL should have provided a re-duction of ;30% in major CV events,close to this data line and to the originallypredicted primary event reduction rate(HR 0.65). The observed 20% reduction inmajor CV events (HR 0.80 [95% CI 0.59–1.07]) is quite similar to the reduction that

would be predicted by SupplementaryFig. 6, given the observed 23.6 mg/dLreduction in LDL-C (after 1 year). In otherwords, the primary results of the EMPATHYstudy, despite not being significant, areconsistent with previous findings. We hy-pothesize that if we had achieved thepredicted difference of 40 mg/dL LDL-Cbetween two treatment groups, the pri-mary end point would have reached sta-tistical significance. This suggests thataggressive LDL-Cmanagementmay furtherreduce risk in patients with conditions suchas hyperlipidemia and diabetic retinopathy.

To further clarify the reasons for thefailure to demonstrate the efficacy of in-tensive therapy, we performed post hocanalysis, classifying patient data into foursubcategories (mean LDL-C ,70, 70 to,100, 100 to ,120, and $120 mg/dLduring the study). Our exploratory find-ings tended to show event prevention atlower LDL-C values in both the intensiveand standard groups (SupplementaryFig. 7). We plan additional exploratoryanalysis of between-group comparison,limited to patients in each group withLDL-C levels within the targeted range.We obtained preliminary results show-ing that the event rate for the primaryend point was significantly lower in theintensive group than in the standard group.

Table 2—Safety issues: AEs and ADRs

Intensive therapy (n = 2,511) Standard therapy (n = 2,518)

Events (n) Patients (n [%]) Events (n) Patients (n [%]) P value

AEsTotal 7,832 1,890 (75.3) 8,189 1,894 (75.2) 0.97Serious 815 535 (21.3) 901 554 (22.0) 0.55

ADRsTotal 368 253 (10.1) 218 168 (6.7) ,0.001Serious 41 32 (1.3) 28 23 (0.9) 0.22

Main AEsHepatobiliary disordersTotal 82 71 (2.8) 52 48 (1.9) 0.03Serious 29 22 (0.9) 14 13 (0.5) 0.13

Renal and urinary disordersTotal 200 166 (6.6) 250 215 (8.5) 0.01Serious 26 21 (0.8) 28 28 (1.1) 0.39

RhabdomyolysisTotal 3 3 (0.1) 4 4 (0.2) 1.00Serious 1 1 (0.0) 1 1 (0.0) 1.00

MyopathyTotal 1 1 (0.0) 0 0 0.50Serious 1 1 (0.0) 0 0 0.50

Cancer*Total 132 114 (4.5) 107 120 (4.2) 0.63Serious 94 81 (3.2) 91 80 (3.2) 0.94

P values were calculated using the Fisher exact test. *Including neoplasms benign, malignant, and unspecified, including cysts and polyps.

care.diabetesjournals.org Itoh and Associates 1281

Our findings from that subanalysis will bepublished separately.In an intervention study where the

control group also receives statin ther-apy, the duration of follow-up is ofconsiderable importance. Non-Japanesestudieswith a relative difference in LDL-C of40–50% between two groups required2 years to show significance in incidencebetween groups (19,20). EMPATHY follow-up time was 3 years, but the between-group difference in LDL-C was lower thananticipated; longer follow-up might haveshown greater between-group difference.Our study data indicated that intensive

therapy favorably affected cerebral events,particularly cerebral infarction. An in-ference of significant reduction cannotbe supported given the lack of a statisticallysignificant difference in the primary endpoint. However, we note that in the StrokePrevention by Aggressive Reduction inCholesterol Levels (SPARCL) study (21),statin therapy effectively reduced cere-brovascular events when LDL-C was low-ered to 70 mg/dL, whereas in the JapanStatin Treatment Against RecurrentStroke (J-STARS) study (22), whereLDL-C levels were reduced only to 100mg/dL, no such efficacy was detected.Exploratory results from EMPATHY weresimilar to those from SPARCL in that LDL-C reached 70mg/dL andmay confirm thata decrease in LDL-C contributes to reduc-ing the event risk for cerebral infarction.Of further interest, the Japan Diabetes

Complications Study (JDCS) (2) reportedthat the incidences of coronary heartdisease and stroke were 9.59/1,000 and7.45/1,000 person-years, respectively, inJapanese patients with type 2 diabetes.In EMPATHY, the incidences of cardiacevents and stroke were 8.13/1,000 and6.14/1,000 person-years, respectively, inthe standard group. The incidences ofcardiac events and ischemic stroke weresurprisingly similar between the JDCS andEMPATHY. This is noteworthy because theEMPATHY study targeted patients whowere already under statin therapy, andthe LDL-C baseline in the standard group(106 mg/dL) was substantially lower thanin the JDCS patients (120 vs. 135 mg/dL).These findings suggest that Japanese pa-tients with diabetic retinopathy may be atparticularly high risk for CV events.No major AE or severe ADR increases

were associatedwith statin monotherapytargeting LDL-C,70 mg/dL, and rhabdo-myolysis occurred at a similar rate in both

groups. This suggested low potential riskfor excessive LDL-C reduction by statinmonotherapy. Aggressive LDL-C loweringmight be associated with increased levelsof cerebral hemorrhage, since cerebralhemorrhage increasedwith intensive ther-apy in the SPARCL study (21). However,subsequent investigation of SPARCLshowed that on-treatment LDL-C didnot predict cerebral hemorrhage (23).EMPATHY findings showed no intergroupdifferences in cerebral hemorrhage, po-tentially eliminating concerns of in-creased cerebral hemorrhage risk due tointensive statin therapy. Although con-cerns have been raised over worsening ofHbA1c due to statin therapy (24,25), nosuch effects were noted in this study.

Key strengths of this study were imple-mentation of the first-ever assessment ofbenefits of intensive statin therapy in pa-tients with hypercholesterolemia and dia-betic retinopathy in primary prevention;suitability assessment of a treat-to-targetapproach for LDL-Cmanagement by titratingstatin therapy; and inclusion of clinics as overhalf the participating sites so that patientmanagement was closer to routine clinicalpractice, which is advantageous for assessingreal-world effects of lipid-lowering therapy.

Limitations of this study were the useof PROBE rather than double-blind com-parison of the two treatment methods;inclusion of soft end points (although thoseend points were assessed by a blindedindependent committee to ensure objec-tivity); unexpectedly high patient discon-tinuation from data loss during a majorearthquake,whichmay have reduced studypower even though discontinuationswerewell balanced between the two groups;and failure to control LDL-C within thetarget range in some patients in bothgroups, resulting in a smaller-than-plannedintergroup difference in LDL-C that possiblyaffected study power. Median LDL-C levelat 6 months in the intensive group was83.0 mg/dL, indicating failure to reach theprotocol-stipulated target LDL-C levelof,70 mg/dL. In real-world clinical prac-tice, many Japanese physicians who arenot lipid management experts worry aboutadverse effects such as intracranial hemor-rhage from intensive LDL-C lowering, andmay have been somehowaffected by theseconcerns even when the protocol stipu-lated the aggressive target of,70 mg/dL.

In conclusion, the intensive therapy tar-geting LDL-C ,70 mg/dL did not reducethe incidence of composite CV events

more significantly than the standard therapytargeting LDL-C $100 and ,120 mg/dL.However, based on previous studies inpatients with diabetes in a primary pre-vention setting, the event reduction ratein this study is consistent with expect-ations regarding the LDL-C difference thatwas achieved andmight be clinicallymean-ingful. Exploratory results suggest that in-tensive therapy reduces cerebral events,especially cerebral infarction, with no in-crease in AEs in this population.

Acknowledgments. EDIT, Inc. (Tokyo, Japan) pro-vided medical writing and editing.Funding andDuality of Interest. This work wassupported by Shionogi & Co., Ltd. H.I. reportsgrants and personal fees from Shionogi & Co., Ltd.during the conducting of the study and grants andpersonal fees from Takeda Pharmaceutical Co.,Ltd., Nippon Boehringer Ingelheim Co., Ltd.,Daiichi Sankyo Co., Ltd., MSD K.K., MitsubishiTanabe Pharma Corp., Shionogi & Co., Ltd., andTaisho Toyama Pharmaceutical Co., Ltd., grantsfrom Sumitomo Dainippon Pharma Co., Ltd.,Astellas Pharma Inc., Kyowa Hakko Kirin Co.,Ltd., Teijin Pharma Ltd., Mochida PharmaceuticalCo., Ltd., Ono Pharmaceutical Co., Ltd., ChugaiPharmaceutical Co., Ltd., and Eli Lilly and Com-pany Japan K.K., and personal fees from NiproCorp. and SBI Pharmaceuticals Co., Ltd. outsidethe submitted work. I.K. reports personal feesfrom Shionogi & Co., Ltd. during the conductingof the study and grants and personal feesfrom Takeda Pharmaceutical Co., Ltd., NipponBoehringer Ingelheim Co., Ltd., Astellas PharmaInc., Daiichi Sankyo Co., Ltd., and Otsuka Pharma-ceutical Co., Ltd. andgrants fromMSDK.K., Shionogi& Co., Ltd., GlaxoSmithKline K.K., Sanofi K.K.,Genzyme Japan K.K., Sumitomo DainipponPharma Co., Ltd., Mitsubishi Tanabe PharmaCorp., and Bristol-Myers Squibb outside the sub-mitted work. M.T. reports personal fees fromShionogi & Co., Ltd. during the conducting ofthe study. T.A. reports personal fees from Shionogi& Co., Ltd. during the conducting of the studyand grants and personal fees from St. JudeMedical Japan Co., Ltd., Terumo Corp., DaiichiSankyo Co., Ltd., and Abbott Vascular Japan Co.,Ltd., grants from Goodman Co., Ltd., OtsukaPharmaceutical Co., Ltd., Pfizer Japan Inc., BayerYakuhin, Ltd., and Boston Scientific Corp., andpersonal fees from Nippon Boehringer IngelheimCo., Ltd. outside the submitted work. H.D. re-ports grants and personal fees from Shionogi &Co., Ltd. during the conducting of the study, andgrants and personal fees from AstraZeneca K.K.,Astellas Pharma Inc., Abbott Vascular Japan Co.,Ltd., Otsuka Pharmaceutical Co., Ltd., Kaken Phar-maceutical Co., Ltd., Kissei Pharmaceutical Co.,Ltd., Kyowa Hakko Kirin Co., Ltd., Kowa Phar-maceutical Company Ltd., Sanofi K.K., DaiichiSankyo Co., Ltd., Sumitomo Dainippon PharmaCo., Ltd., Takeda Pharmaceutical Co., Ltd., Ter-umo Corp., Nippon Boehringer Ingelheim Co.,Ltd., Bayer Yakuhin, Ltd., Pfizer Japan Inc.,Philips Respironics GK, Bristol-Myers Squibb,Sanwa Kagaku Kenkyusho Co., Ltd., Mitsubishi

1282 Intensive Statin Therapy in High-Risk Patients Diabetes Care Volume 41, June 2018

Tanabe Pharma Corp., MSD K.K., and GlaxoS-mithKline K.K., grants from Eisai Co., Ltd., TeijinPharma Ltd., Nippon Shinyaku Co., Ltd., VitalAireJapan K.K., Fujifilm RI Pharma Co., Ltd., BostonScientific Corp., Fuji Chemical Industries Co.,Ltd., Fukuda Denshi Co., Ltd., and Actelion Phar-maceuticals Japan Ltd., and personal fees fromASKA Pharmaceutical Co., Ltd., Chugai Pharma-ceutical Co., Ltd., Taisho Toyama PharmaceuticalCo., Ltd., Toa Eiyo Ltd., Ono Pharmaceutical Co.,Ltd., Medtronic Japan Co., Ltd., and MochidaPharmaceutical Co., Ltd. outside the submittedwork. Y.E. reports nonfinancial support fromShionogi & Co., Ltd. during the conducting ofthe study. H.F. reports other fees (consultant)from Mehergen Group Holdings, Inc. outside thesubmitted work. J.H. reports grants and personalfees from Shionogi & Co., Ltd. during the con-ducting of the study and grants and personal feesfrom Astellas Pharma Inc., Nippon BoehringerIngelheim Co., Ltd., Mochida PharmaceuticalCo., Ltd., Daiichi Sankyo Co., Ltd., Takeda Phar-maceutical Co., Ltd., Sumitomo DainipponPharma Co., Ltd., MSD K.K., Teijin Pharma Ltd.,Actelion Pharmaceuticals Japan Ltd., Otsuka Phar-maceutical Co., Ltd., Novartis Pharma K.K., andSanwa Kagaku Kenkyusho Co., Ltd. outside thesubmitted work. K.-i.H. reports personal fees andnonfinancial support from Shionogi & Co., Ltd.during the conducting of the study and grants andpersonal fees from Daiichi Sankyo Co., Ltd. andMochida Pharmaceutical Co., Ltd., grants fromActelion Pharmaceuticals Japan Ltd., Eisai Co.,Ltd., Otsuka Pharmaceutical Co., Ltd., SumitomoDainippon Pharma Co., Ltd., Takeda Pharmaceu-tical Co. Ltd., Nippon Boehringer Ingelheim Co.,Ltd., Bayer Yakuhin, Ltd., Sysmex Corp., Med-tronic Japan Co., Ltd., and St. Jude Medical JapanCo., Ltd., and personal fees from Kowa Pharma-ceutical Co., Ltd. outside the submittedwork. S.Is.reports grants and personal fees from Shionogi &Co., Ltd. during the conducting of the study andgrants and personal fees from Amgen AstellasBioPharma K.K., Astellas Pharma Inc., DaiichiSankyo Co., Ltd., Eli Lilly and Company JapanK.K., Kowa Pharmaceutical Co., Ltd., NipponBoehringer Ingelheim Co., Ltd., Kissei Pharmaceu-tical Co., Ltd., MSD K.K., Novartis Pharma K.K.,Mitsubishi Tanabe Pharma Corp., Ono Pharma-ceutical Co. Ltd., Sanofi K.K., Takeda Pharmaceu-tical Co., Ltd., Taisho Toyama Pharmaceutical Co.,Ltd., and Teijin Pharma Ltd., grants from FujifilmPharma Co., Ltd., Sumitomo Dainippon PharmaCo., Ltd., and Kyowa Hakko Kirin Co., Ltd., andpersonal fees from AstraZeneca K.K., BayerYakuhin, Ltd., Novo Nordisk Pharma Ltd., PfizerJapan Inc., and Sanwa Kagaku Kenkyusho Co., Ltd.outside the submitted work. T.I. reports personalfees and nonfinancial support from Shionogi &Co., Ltd. during the conducting of the studyand grants and personal fees from Sanofi K.K.,Sumitomo Dainippon Pharma Co., Ltd., and DaiichiSankyo Co., Ltd., grants from Takeda Pharmaceu-tical Co., Ltd. and Mitsubishi Tanabe PharmaCorp., and personal fees from Astellas PharmaInc., AstraZeneca K.K., and MSD K.K. outside thesubmitted work. S.It. reports grants, personalfees, and nonfinancial support from Shionogi &Co., Ltd. during the conducting of the study. A.K.reports personal fees and nonfinancial supportfrom Shionogi & Co., Ltd. during the conductingof the study and personal fees from Astellas

Pharma Inc., Sunstar Group Ltd., Eli Lilly andCompany Japan K.K., Sanofi K.K., AstraZenecaK.K., Takeda Pharmaceutical Co., Ltd., TaishoToyama Pharmaceutical Co., Ltd., Nippon BoehringerIngelheim Co., Ltd., Kowa Pharmaceutical Co.,Ltd., and Sanwa Kagaku Kenkyusho Co., Ltd. out-side the submitted work. S.K. reports grants fromShionogi & Co., Ltd. during the conducting of thestudy. K.K. reports grants and personal fees fromShionogi & Co., Ltd. during the conducting of thestudy. M.Ki. reports grants and personal feesfrom Shionogi & Co., Ltd. during the conductingof the study and grants and personal fees fromAstellas Pharma Inc., Sanofi K.K., Pfizer Japan Inc.,Ono Pharmaceutical Co., Ltd., Novartis PharmaK.K., Mitsubishi Tanabe Pharma Corp., KyowaHakko Kirin Co., Ltd., Abbott Japan Co., Ltd.,and Otsuka Pharmaceutical Co., Ltd., grantsfrom the Japanese government, Japan HeartFoundation, Japan Cardiovascular Research Foun-dation, Calpis Co., Ltd., and Nihon Kohden Corp.,and personal fees from Daiichi Sankyo Co., Ltd.,Bayer Yakuhin Ltd., Nippon Boehringer IngelheimCo., Ltd., Kowa Pharmaceutical Co., Ltd., SumitomoDainippon Pharma Co., Ltd., Sawai Pharma-ceutical Co., Ltd., MSD K.K., Shionogi & Co.,Ltd., AstraZeneca K.K., Asahi Kasei Medical Co.,Ltd., Novo Nordisk Pharma Ltd., Fujifilm RIPharma Co., Ltd., and Japan Medical Data outsidethe submitted work. T.K. reports grants andpersonal fees from Shionogi & Co., Ltd. duringthe conducting of the study and grants andpersonal fees from Daiichi Sankyo Co., Ltd. andBayer Yakuhin Ltd. and grants from MSD K.K.,Novartis Pharma K.K., Astellas Pharma Inc., andPfizer Japan Inc. outside the submitted work. M.Ku. reports personal fees from Shionogi & Co.,Ltd. during the conducting of the study and grantsand personal fees from Shionogi & Co., Ltd.outside the submitted work. K.M. reports other(meeting attendance fee) from Shionogi & Co.,Ltd. during the conducting of the study. T.Mura.reports personal fees from Shionogi & Co., Ltd.during the conducting of the study. T.Muro.reports personal fees from Shionogi & Co., Ltd.during the conducting of the study and grants andpersonal fees from Daiichi Sankyo Co., Ltd., PfizerJapan Inc., Kowa Pharmaceutical Co., Ltd., MSDK.K., and Mitsubishi Tanabe Pharma Corp. andpersonal fees from AstraZeneca K.K. outside thesubmitted work. K.No. reports nonfinancial sup-port from Shionogi & Co., Ltd. during the con-ducting of the study. S.O. reports personal feesand nonfinancial support from Shionogi & Co.,Ltd. during the conducting of the study. Y.Sa.reports grants, personal fees, and nonfinancialsupport from Shionogi & Co., Ltd. during theconducting of the study and grants, personalfees, and other (advisory boards) from MSDK.K., Ono Pharmaceutical Co., Ltd., MitsubishiTanabe Pharma Corp., Pfizer Japan Inc., andNovartis Pharma K.K., grants and personal feesfrom Daiichi Sankyo Co., Ltd., Bayer Yakuhin,Ltd., Otsuka Pharmaceutical Co., Ltd., SumitomoDainippon Pharma Co., Ltd., Astellas Pharma Inc.,and Takeda Pharmaceutical Co., Ltd., and grantsfrom Baxter Ltd., Kyowa Hakko Kirin Co., Ltd.,Teijin Pharma Ltd., Eisai Co., Ltd., Zeria Pharma-ceutical Co., Ltd., Nihon Medi-Physics Co., Ltd.,Chugai Pharmaceutical Co., Ltd., Genzyme JapanK.K., and Medtronic Japan Co., Ltd. outside thesubmitted work. Y.Se. reports personal fees from

Shionogi & Co., Ltd. during the conducting of thestudy and grants and personal fees from OtsukaPharmaceutical Co., Ltd. and Nippon BoehringerIngelheim Co., Ltd. and grants from MitsubishiTanabe Pharma Corp., Fujifilm RI Pharma Co.,Ltd., Roche Diagnostics K.K., MSD K.K., PfizerJapan Inc., Bayer Yakuhin, Ltd., and Shionogi &Co., Ltd. outside the submitted work. T.S. reportspersonal fees and nonfinancial support fromShionogi & Co., Ltd. during the conducting ofthe study. S.Sh. reports personal fees and non-financial support from Shionogi & Co., Ltd. duringthe conducting of the study. M.S. reports per-sonal fees and nonfinancial support fromShionogi & Co., Ltd. during the conducting ofthe study. S.Su. reports personal fees from Shio-nogi & Co., Ltd. during the conducting of thestudy and grants from the Ministry of Education,Culture, Sports, Science and Technology in Japanoutside the submitted work. Y.T. reports personalfees from Shionogi & Co., Ltd. during the con-ducting of the study and grants and personal feesfrom Astellas Pharma Inc., AstraZeneca K.K., BayerYakuhin, Ltd., Daiichi Sankyo Co., Ltd., SumitomoDainippon Pharma Co., Ltd., Eli Lilly and CompanyJapan K.K., Kissei Pharmaceutical Co., Ltd., KowaPharmaceutical Co., Ltd., Kyowa Hakko Kirin Co.,Ltd., MSD K.K., Mitsubishi Tanabe Pharma Corp.,Nippon Boehringer Ingelheim Co., Ltd., NovoNordisk Pharma Ltd., Ono Pharmaceutical Co.,Ltd., Sanwa Kagaku Kenkyusho Co., Ltd., SanofiK.K., Shionogi & Co., Ltd., Taisho Toyama Phar-maceutical Co., Ltd., and Takeda PharmaceuticalCo., Ltd. and personal fees from Novartis PharmaK.K. outside the submitted work. H.T. reportspersonal fees and nonfinancial support fromShionogi & Co., Ltd. during the conducting ofthe study and grants and personal fees fromDaiichi Sankyo Co., Ltd. and Takeda Pharmaceu-tical Co., Ltd., grants from Novartis Pharma K.K.and Astellas Pharma Inc., and personal fees fromMSD K.K., Otsuka Pharmaceutical Co., Ltd., PfizerJapan Inc., Mitsubishi Tanabe Pharma Corp.,Teijin Pharma Ltd., Nippon Boehringer IngelheimCo., Ltd., Bayer Yakuhin, Ltd., and Bristol-MyersSquibb outside the submitted work. K.Ue. reportsother (contracted work) from Shionogi & Co., Ltd.during the conducting of the study and personalfees from Shionogi & Co., Ltd. outside the sub-mitted work. K.Ut. reports personal fees andnonfinancial support from Shionogi & Co., Ltd.during the conducting of the study and grantsfrom Sanofi K.K., MSD K.K., Taisho Toyama Phar-maceutical Co., Ltd., Nippon Boehringer lngel-heim Co., Ltd., Takeda Pharmaceutical Co., Ltd.,Eli Lilly and Company Japan K.K., and NovoNordisk Pharma Ltd. outside the submittedwork. M.Ya. reports personal fees from Shionogi& Co., Ltd. during the conducting of the study andother (donation) from Shionogi & Co., Ltd. out-side the submitted work. T.Y. reports other (lec-ture fee) from Shionogi & Co., Ltd. during theconducting of the study. S.Y. reports other (con-tractedwork) from Shionogi & Co., Ltd. during theconducting of the study. K.Yok. reports personalfees from Shionogi & Co., Ltd. during the con-ducting of the study and grants, personal fees,and nonfinancial support from MSD K.K., grantsand personal fees from Astellas Pharma Inc.,Daiichi Sankyo Co., Ltd., Sumitomo DainipponPharma Co., Ltd., Kyowa Hakko Kirin Co., Ltd.,Mochida Pharmaceutical Co., Ltd., Nippon

care.diabetesjournals.org Itoh and Associates 1283

Boehringer lngelheim Co., Ltd., Ono Pharmaceu-tical Co., Ltd., Pfizer Japan Inc., Shionogi & Co.,Ltd., Taisho Toyama Pharmaceutical Co., Ltd.,Takeda Pharmaceutical Company Ltd., andMitsubishiTanabe Pharma Corp., grants from Bristol-MyersSquibb, Eli Lilly and Company Japan K.K., TeijinPharma Ltd., and Toyama Chemical Co., Ltd., andpersonal fees from AstraZeneca K.K., Eisai Co.,Ltd., Kowa Company, Ltd., Kowa PharmaceuticalCo., Ltd., Sanofi K.K., and Sanwa Kagaku Ken-kyusho Co., Ltd. outside the submitted work. K.Yos. reports personal fees and nonfinancial sup-port from Shionogi & Co., Ltd. during the con-ducting of the study. M.Yo. reports grants andpersonal fees from Shionogi & Co., Ltd. outsidethe submitted work. N.Y. reports personal feesfrom Shionogi & Co., Ltd. during the conductingof the study and personal fees from Shionogi &Co., Ltd. outside the submitted work. K.Na. re-ports other (contracted) work from Shionogi &Co., Ltd. during the conducting of the study andgrants from Takeda Pharmaceutical Co., Ltd. andFujifilm Pharma Co., Ltd. outside the submittedwork. R.N. reports personal fees from Shionogi &Co., Ltd. during the conducting of the study andpersonal fees from Astellas Pharma Inc., Sumi-tomo Dainippon Pharma Co., Ltd., MSD K.K., OnoPharmaceutical Co. Ltd., Kowa Pharmaceutical Co.,Ltd., Mitsubishi Tanabe Pharma Corp., NipponBoehringer Ingelheim Co., Ltd., Toa Eiyo Ltd., EisaiCo., Ltd., and Nippon Chemiphar Co., Ltd. outsidethe submitted work. No other potential conflicts ofinterest relevant to this article were reported.Author Contributions. H.I., I.K., H.F., T.Muro.,and T.Y. designed the study, collected and inter-preted data, and contributed to the writing. M.T.collected and analyzed data and contributed tothe writing. T.A., J.H., T.I., A.K., M.Ki., T.K., M.Ku.,K.No., S.O., Y.Sa., Y.Se., T.S., S.Sh., and S.Y. in-terpreted data. H.D. and K.Ut. designed the studyand collected and interpreted data. Y.E. collecteddata. K.-i.H., S.It., S.Su., K.Yok., and K.Na. designedthe study and interpreted data. S.Is., K.K., Y.T., andM.Ya. collected and interpreted data and contrib-uted to the writing. S.K. designed the study, col-lected data, and contributed to the writing. K.M.designed the study and collected data. T.Mura.designed the study. M.S. collected data and con-tributed to the writing. H.T. and N.Y. interpreteddata and contributed to the writing. K.Ue. and R.N.designed the study, interpreted data, and contrib-uted to the writing. K.Yos. and M.Yo. collected andinterpreted data. All authors read the drafted man-uscript, provided feedback, and approved the finalversion of the manuscript. H.I. is the guarantor ofthis work and, as such, had full access to all the datain the study and takes responsibility for the integrityof the data and the accuracy of the data analysis.Prior Presentation. This study was presented atthe European Society of Cardiology Congress 2017Hot Line session, Barcelona, Spain, 29 August 2017.

References1. Turner RC, Millns H, Neil HA, et al. Risk factorsfor coronary artery disease in non-insulin dependentdiabetes mellitus: United Kingdom Prospective Di-abetes Study (UKPDS: 23). BMJ 1998;316:823–828

2. Sone H, Tanaka S, Tanaka S, et al.; JapanDiabetes Complications Study Group. Serum levelof triglycerides is a potent risk factor comparableto LDL cholesterol for coronary heart disease inJapanese patients with type 2 diabetes: subanal-ysis of the Japan Diabetes Complications Study(JDCS). J Clin Endocrinol Metab 2011;96:3448–34563. Gaede P, Pedersen O. Intensive integrated ther-apy of type 2 diabetes: implications for long-termprognosis. Diabetes 2004;53(Suppl. 3):S39–S474. Kearney PM, Blackwell L, Collins R, et al.;Cholesterol Treatment Trialists’ (CTT) Collaborators.Efficacy of cholesterol-lowering therapy in 18,686people with diabetes in 14 randomised trials ofstatins: a meta-analysis. Lancet 2008;371:117–1255. Kramer CK, Rodrigues TC, Canani LH, Gross JL,Azevedo MJ. Diabetic retinopathy predicts all-cause mortality and cardiovascular events inboth type 1 and 2 diabetes: meta-analysis ofobservational studies. Diabetes Care 2011;34:1238–12446. Ohno T, Kinoshita O, Fujita H, et al. Detectingoccult coronary artery disease followed by earlycoronary artery bypass surgery in patients withdiabetic retinopathy: report from a diabetic ret-inocoronary clinic. J Thorac Cardiovasc Surg 2010;139:92–977. Baigent C, Blackwell L, Emberson J, et al.;Cholesterol Treatment Trialists’ (CTT) Collabora-tion. Efficacy and safety of more intensive low-ering of LDL cholesterol: a meta-analysis of datafrom 170,000 participants in 26 randomised trials.Lancet 2010;376:1670–16818. Stone NJ, Robinson JG, Lichtenstein AH, et al.;American College of Cardiology/American HeartAssociation Task Force on Practice Guidelines.2013 ACC/AHA guideline on the treatment ofblood cholesterol to reduce atherosclerotic car-diovascular risk in adults: a report of the AmericanCollege of Cardiology/American Heart AssociationTask Force on Practice Guidelines [published cor-rection appears in J Am Coll Cardiol 2014;63:3024–3025, 2015;66:2812]. J Am Coll Cardiol 2014;63(25 Pt. B):2889–29349. Ueshima K, Itoh H, Kanazawa N, et al.;EMPATHY study group. Rationale and design ofthe standard versus intEnsive statin therapy forhypercholesteroleMic Patients with DiAbeticRetinopaTHY (EMPATHY) study: a randomizedcontrolled trial. J Atheroscler Thromb 2016;23:976–99010. Brunzell JD, Davidson M, Furberg CD, et al.;American Diabetes Association; American Collegeof Cardiology Foundation. Lipoprotein manage-ment in patients with cardiometabolic risk: con-sensus statement from the American DiabetesAssociation and the American College of Cardi-ology Foundation. Diabetes Care 2008;31:811–82211. Teramoto T, Sasaki J, Ueshima H, et al. Ex-ecutive summary of Japan Atherosclerosis Society(JAS) guideline for diagnosis and prevention ofatherosclerotic cardiovascular diseases for Japa-nese. J Atheroscler Thromb 2007;14:45–5012. Sandhu S, Wiebe N, Fried LF, Tonelli M.Statins for improving renal outcomes: a meta-analysis. J Am Soc Nephrol 2006;17:2006–2016

13. Colhoun HM, Betteridge DJ, Durrington PN,et al.; CARDS investigators. Primary preventionof cardiovascular disease with atorvastatin intype 2 diabetes in the Collaborative AtorvastatinDiabetes Study (CARDS): multicentre randomisedplacebo-controlled trial. Lancet 2004;364:685–69614. Sever PS, Poulter NR, Dahlof B, et al. Re-duction in cardiovascular events with atorvas-tatin in 2,532 patients with type 2 diabetes:Anglo-Scandinavian Cardiac Outcomes Trial–lipid-lowering arm (ASCOT-LLA). Diabetes Care2005;28:1151–115715. Collins R, Armitage J, Parish S, Sleigh P, PetoR; Heart Protection Study Collaborative Group.MRC/BHF Heart Protection Study of cholesterol-lowering with simvastatin in 5963 people withdiabetes: a randomised placebo-controlled trial.Lancet 2003;361:2005–201616. Ogawa H, Nakayama M, Morimoto T, et al.;Japanese Primary Prevention of AtherosclerosisWith Aspirin for Diabetes (JPAD) Trial Investiga-tors. Low-dose aspirin for primary preventionof atherosclerotic events in patients with type 2diabetes: a randomized controlled trial. JAMA2008;300:2134–214117. Cannon CP, Blazing MA, Giugliano RP, et al.;IMPROVE-IT Investigators. Ezetimibe added tostatin therapy after acute coronary syndromes.N Engl J Med 2015;372:2387–239718. Baigent C, Keech A, Kearney PM, et al.;Cholesterol Treatment Trialists’ (CTT) Collabora-tors. Efficacy and safety of cholesterol-loweringtreatment: prospective meta-analysis of datafrom 90,056 participants in 14 randomisedtrials of statins. Lancet 2005;366:1267–127819. Sabatine MS, Giugliano RP, Keech AC, et al.;FOURIER Steering Committee and Investigators.Evolocumab and clinical outcomes in patientswith cardiovascular disease. N Engl J Med 2017;376:1713–172220. Bowman L, Hopewell JC, Chen F, et al.; HPS3/TIMI55–REVEAL Collaborative Group. Effects ofanacetrapib in patients with atherosclerotic vas-cular disease. N Engl J Med 2017;377:1217–122721. Amarenco P, Bogousslavsky J, Callahan A 3rd,et al.; Stroke Prevention by Aggressive Reductionin Cholesterol Levels (SPARCL) Investigators.High-dose atorvastatin after stroke or transientischemic attack. N Engl J Med 2006;355:549–55922. HosomiN, Nagai Y, KohriyamaT, et al.; J-STARSCollaborators. The Japan Statin Treatment AgainstRecurrent Stroke (J-STARS): a multicenter, ran-domized, open-label, parallel-group study. EBio-Medicine 2015;2:1071–107823. Goldstein LB, Amarenco P, Szarek M, et al.;SPARCL Investigators. Hemorrhagic stroke in theStroke Prevention by Aggressive Reduction in Choles-terol Levels study. Neurology 2008;70:2364–237024. Rajpathak SN, Kumbhani DJ, Crandall J,Barzilai N, AldermanM, Ridker PM. Statin therapyand risk of developing type 2 diabetes: a meta-analysis. Diabetes Care 2009;32:1924–192925. Sattar N, Preiss D, Murray HM, et al. Statinsand risk of incident diabetes: a collaborative meta-analysis of randomised statin trials. Lancet 2010;375:735–742

1284 Intensive Statin Therapy in High-Risk Patients Diabetes Care Volume 41, June 2018