538 Diabetes Care Volume 41, March 2018

fi Eiichi Araki,1 Shizuya Yamashita,2,3 Effects of Pema brate, a Novel Hidenori Arai,4 Koutaro Yokote,5 Jo Satoh,6 Toyoshi Inoguchi,7 Jiro Nakamura,8 Selective PPARa Modulator, on Hiroshi Maegawa,9 Narihito Yoshioka,10 Yukio Tanizawa,11 Hirotaka Watada,12 Lipid and Glucose Metabolism in Hideki Suganami,13 and Shun Ishibashi14 Patients With Type 2 Diabetes and Hypertriglyceridemia: A Randomized, Double-Blind, 1Department of Metabolic Medicine, Faculty of Placebo-Controlled, Phase 3 Trial Life Sciences, Kumamoto University, Kumamoto, Japan – Diabetes Care 2018;41:538 546 | https://doi.org/10.2337/dc17-1589 2Department of Community Medicine and De- partment of Cardiovascular Medicine, Osaka University Graduate School of Medicine, Osaka, Japan 3Rinku General Medical Center, Osaka, Japan 4National Center for Geriatrics and Gerontology, Aichi, Japan 5Department of Clinical Cell Biology and Medi- cine, Chiba University Graduate School of Medi- cine, Chiba, Japan 6Tohoku Medical and Pharmaceutical University Wakabayashi Hospital, Miyagi, Japan 7Fukuoka Health Promotion Support Center, OBJECTIVE Fukuoka, Japan Type 2 diabetes is frequently complicated with atherogenic dyslipidemia. This study 8Division of Diabetes, Department of Internal aimed to evaluate the efficacy and safety of pemafibrate (K-877) in patients with Medicine, Aichi Medical University, Aichi, Japan 9Department of Medicine, Shiga University of type 2 diabetes comorbid with hypertriglyceridemia. Medical Science, Shiga, Japan 10Division of Diabetes and Endocrinology, De- RESEARCH DESIGN AND METHODS partment of Medicine, Sapporo Medical Center, Patients were randomly assigned to three groups and received placebo (n =57), NTT East Corporation, Hokkaido, Japan 11 0.2 mg/day pemafibrate (n = 54), or 0.4 mg/day pemafibrate (n = 55) for 24 weeks Division of Endocrinology, Metabolism, Hema- tological Science and Therapeutics, Yamaguchi (treatment period 1). Subsequently, the patients received follow-up treatment University Graduate School of Medicine, Yama- for another 28 weeks (treatment period 2), in which the placebo was switched to guchi, Japan 12 EMERGING TECHNOLOGIES AND THERAPEUTICS 0.2 mg/day pemafibrate. This article presents the results of treatment period 1, Department of Metabolism and Endocrinol- which were the primary objectives. ogy, Juntendo University Graduate School of Medicine, Tokyo, Japan 13 RESULTS Clinical Data Science Department, Kowa Com- pany, Ltd., Tokyo, Japan The pemafibrate groups showed significantly reduced fasting serum triglyceride 14Division of Endocrinology and Metabolism, levels by ∼45% compared with the placebo group (P < 0.001). Additionally, the Department of Medicine, Jichi Medical Univer- pemafibrate groups displayed significant decreases in non-HDL and remnant lipo- sity, Tochigi, Japan protein cholesterol, apolipoprotein (Apo) B100, ApoB48, and ApoCIII levels and sig- Corresponding author: Eiichi Araki, earaki@gpo .kumamoto-u.ac.jp. nificant increases in HDL cholesterol and ApoA-I levels. LDL cholesterol levels were fi Received 1 August 2017 and accepted 27 No- not considerably altered in the pema brate groups. Furthermore, the 0.2 mg/day vember 2017. fi fi – pema brate group showed a signi cantly reduced HOMA insulin resistance score Clinical trial reg. no. JapicCTI-142412, clinicaltrials compared with the placebo group; however, no significant changes compared with .jp. placebo were found in fasting plasma glucose, fasting insulin, glycoalbumin, or HbA1c This article contains Supplementary Data online levels. The pemafibrate groups also showed significantly increased fibroblast growth at http://care.diabetesjournals.org/lookup/ factor 21 levels compared with the placebo group. All groups displayed comparable suppl/doi:10.2337/dc17-1589/-/DC1. rates of adverse events and drug reactions. © 2018 by the American Diabetes Association. Readers may use this article as long as the work CONCLUSIONS is properly cited, the use is educational and not for profit, and the work is not altered. More infor- Pemafibrate significantly ameliorated lipid abnormalities and was well tolerated in mation is available at http://www.diabetesjournals patients with type 2 diabetes comorbid with hypertriglyceridemia. .org/content/license. care.diabetesjournals.org Araki and Associates 539

The leading cause of death in patients with Pemafibrate (K-877) is a novel selective patients provided written informed con- type 2 diabetes is atherosclerotic cardio- peroxisome proliferator–activated receptor sent prior to their participation. This study vascular disease (ASCVD) (1). Cardiovascu- a (PPARa) modulator approved for the is registered at Japan Pharmaceutical In- lar events are more common in patients treatment of dyslipidemia. A dose-finding formation Center Clinical Trials Informa- with diabetes than in those without (2,3). phase 2 trial on pemafibrate conducted in tion (JapicCTI-142412). The increased ASCVD risk in patients with patients with atherogenic dyslipidemia re- Patients diabetes is attributed to abnormalities in vealed that this drug exerted significant TG The inclusion criteria were as follows: 1) both glucose and lipid metabolism. Lipid reduction and HDL-C increase, with compa- type 2 diabetes comorbid with hypertri- abnormalities are often comorbid with rable rates of adverse events (AEs) to pla- glyceridemia ($6.2% [44.3 mmol/mol] type 2 diabetes and are unique in terms cebo, such as serum creatinine and liver HbA and $150 mg/dL [1.7 mmol/L] of quantitative and qualitative lipid ab- enzyme increases, which suggests that pe- 1c fasting serum TG levels for two consecu- normalities that are associated with insu- mafibrate may have a better benefit/ tive screening visits); 2)age$20 years; 3) lin resistance (IR) (4). The clinical features risk balance than fenofibrate (12) men and postmenopausal women; and of these abnormalities include elevated Given that type 2 diabetes is frequently 4) $12 weeks of dietary or exercise guid- triglyceride (TG) levels, reduced HDL cho- complicated with atherogenic dyslipi- ance before the first screening test. lesterol (HDL-C) levels, and delayed TG-rich demia, a considerable proportion of The exclusion criteria were as follows: 1) lipoprotein catabolism, leading to elevated patients treated with pemafibrate are fasting serum TG levels .1,000 mg/dL postprandial TG levels, remnant lipopro- anticipated to have type 2 diabetes. How- (11.3 mmol/L); 2) type 1 diabetes, in- tein accumulation, and increased small ever, to date, the efficacy and safety of adequately controlled diabetes ($8.0% dense LDL production. pemafibrate, specifically in patients with [63.9 mmol/mol] HbA ), diabetes requir- A number of large-scale clinical trials type 2 diabetes comorbid with hypertri- 1c ing treatment with insulin, thiazolidinedi- have demonstrated that the management glyceridemia, have not been investigated ones, biguanides, high-dose sulfonylureas of dyslipidemia resulted in significantly re- through a prospective randomized trial. ($4 mg/day glimepiride, $7.5 mg/day duced cardiovascular risk in patients with Therefore, we conducted this phase 3 clinical glibenclamide, and $120 mg/day gli- diabetes. The Collaborative Atorvastatin trial to evaluate them through placebo- clazide), sodium–glucose cotransporter 2 Diabetes Study (CARDS) (5) and Choles- controlled treatment for 24 weeks (treat- inhibitors, or combination therapy with terol Treatment Trialists (CTT) Collabora- ment period 1), followed by further three or more antidiabetic agents, and re- tion meta-analysis (6) reported that LDL long-term treatment for another 28 weeks, cent changes in the class and dosage of cholesterol (LDL-C)–lowering therapy in which placebo was switched to pemafi- antidiabetic agents within 12 weeks pri- with statins reduces cardiovascular risk brate (treatment period 2). This article is or to the first screening test; 3) inade- in patients with type 2 diabetes. Addition- based on the clinical study report for treat- quately controlled thyroid disorders; 4) ally, the Japan Diabetes Complications ment period 1, which was documented inadequately controlled hypertension Study (JDCS) identified both high LDL-C before the completion of a 52-week treat- ($180/$110 mmHg systolic/diastolic andTGlevelsasriskfactorsforthedevel- ment period. The overall results through- blood pressure); 5) $1.5 mg/dL serum opment of coronary artery disease (7). Fur- out the 52 weeks will be separately creatinine levels for patients receiving thermore, the UK Prospective Diabetes documented in another article based on statin treatment; 6) a creatine kinase Study (UKPDS) revealed that both high the other clinical study report that includes (CK) level that is more than five times LDL-C and low HDL-C levels were associ- the results of treatment period 2. the upper limit of normal (ULN) for pa- ated with elevated coronary artery dis- tients on statin treatment; 7) aspartate ease risk (8). Therefore, research in RESEARCH DESIGN AND METHODS aminotransferase (AST) and alanine ami- recent decades has focused on interven- Study Design notransferase (ALT) levels that are more tions targeting diabetic lipid abnormalities This multicenter, placebo-controlled, than three times the ULN, or serious he- other than high LDL-C levels. In particular, randomized, double-blind, parallel-group patic disorders; 8) gallstones or serious large-scale clinical trials have revealed that study was performed in 34 medical insti- biliary disorders; 9) occurrence of acute treatment with fibrates, which decrease tutions in Japan from 20 February 2014 myocardial infarction or stroke within TG levels and increase HDL-C levels, re- to 30 April 2015, denoted as treatment 3 months before providing informed con- duces ASCVD risk in patients with type 2 period 1. sent; and 10) heart failure with New York diabetes. For example, in the Fenofibrate The study protocol was approved by Heart Association class III or IV. Intervention and Event Lowering in Diabe- the institutional review boards of the tes(FIELD)andActiontoControlCardio- 34 medical institutions prior to the imple- Procedures vascular Risk in Diabetes (ACCORD) Lipid mentation of the study. Additional mat- Patients who provided their informed studies, treatment with fenofibrate led to ters that needed to be approved, such as consent were assessed for eligibility as event reduction in subgroup patients with protocol amendments, were assessed participants (Supplementary Fig. 1). Eligi- high TG and low HDL-C levels (9,10). More- and approved by the institutional review ble patients were randomly and equally over, the post hoc analysis of the Veter- boards as needed. The study was con- assigned to the placebo, 0.2 mg/day ans Affairs High-Density Lipoprotein ducted in accordance with the princi- pemafibrate (twice daily), and 0.4 mg/day Intervention Trial (VA-HIT) showed that ples of the Declaration of Helsinki and pemafibrate (twice daily) groups using treatment with gemfibrozil resulted in the Ministerial Ordinance on Good Clini- the dynamic allocation method with the cardiovascular event reduction in a sub- cal Practice for Drugs issued by the Min- HbA1c value ($7.4% [57.4 mmol/mol] group of patients with diabetes (11). istry of Health and Welfare, Japan. All or ,7.4%) from the second screening test 540 Pemafibrate and Diabetic Dyslipidemia Diabetes Care Volume 41, March 2018

and antidiabetic treatment (no agent, after 12 weeks at Skylight Biotech values. The safety analyses were based on sulfonylurea, or the other antidiabetic to examine the lipoprotein profiles by the safety analysis set. The safety analysis agent) as adjustment factors. An indepen- subclass. set included all patients who received at dent third party generated the random A meal tolerance test was performed at least one dose of the study drug. The full allocation key codes, confirmed that pla- the study sites where this test was feasible analysis set included patients in the safety cebo was indistinguishable from pemafi- at weeks 0 and 24 in patients who provided analysis set who had valid baseline and brate, and conducted the numbering and writteninformedconsent for undergoing postbaseline efficacy measurements. concealment of the study drugs. Another this test. Fasting blood samples were col- The primary efficacy analysis was con- independent third party managed the dy- lected before the meal and study drug ad- ducted using ANCOVA with the baseline namic allocation based on the screening ministration. The patients had the test as a covariate. Multiplicity was adjusted test results. Investigators and participants meal within 15 min in principle and took using the Dunnett test for the comparison were not provided with any information the study drug at 30 min before or after of the effect between the placebo group related to the randomization throughout starting the meal. Postprandial blood sam- and each of the pemafibrate groups. The the study period. ples were collected 0.5, 1, 2, 2.5, 4.5, and secondary efficacy end points were ana- Patients were instructed to take the 6.5 h after starting the meal. The test meal lyzed using a one-sample t test for the assigned study drug twice daily before was Meal Test C (Saraya Co., Ltd., Osaka, differences from the baseline and ANCOVA or after a meal (fixed throughout the Japan), which contained 592 kcal, 28.5 g fat for the differences between groups. The study period) in the morning and evening (derived from butter), 75.0 g carbohydrates primary safety end points were analyzed for 24 weeks. Throughout the study pe- (derived from wheat starch and maltose), using the Fisher exact test. The significance riod, the following drugs were prohibited 8.0 g protein, 0.5–4.0 g dietary fiber, 125 level was 0.05 for a two-sided test. SAS for concomitant medications: fibrates, mg sodium, and 0 g sucrose. version 9.3 (SAS Institute, Inc., Cary, NC) bile acid sequestrants, adrenocorticoste- was used for the analyses. All primary effi- roids (systemic use), and sodium–glucose End Points cacy and safety end point analyses were The primary efficacy end point was the cotransporter 2 inhibitors. In principle, performed based on a prespecified statis- percentage change in fasting serum TG initiation, discontinuation, or change in tical analysis plan. level from the baseline at the final evalu- the dosage regimen was prohibited for any ation over 24 weeks. The secondary effi- antidyslipidemic agent not mentioned as RESULTS cacy end points were the percentage prohibited concomitant medication from Supplementary Figure 2 shows the dispo- changes or changes in the levels of fasting 4 weeks before the screening tests. Such sition of the patients. Among 306 patients and postprandial lipid-related and glyce- changes in the use of antidiabetic agents, who provided written consent, 167 pa- mic parameters from baseline except for protease inhibitors, anabolic steroid hor- tients were eligible and randomly assigned the primary efficacy end point. mones, and progestogen were prohibited. to the three groups. One patient discontin- The primary safety end points were the However, changes in the class and dosage ued participation in the study before study incidence rates of AEs and adverse drug regimen of antidiabetic agents were permit- drug administration because of an AE. reactions (ADRs) after the study drug ad- ted for the next hospital visit (from week 16 Thus, 57, 54, and 55 patients received the ministration. AEs were defined as any un- onward) to treat deterioration in glycemic placebo, 0.2 mg/day pemafibrate, and desirable or unintended signs, symptoms, control found after week 12. Patients with a 0.4 mg/day pemafibrate, respectively. Af- and disorders, including laboratory test drinking habit were instructed to limit alco- ter the study drug administration, the abnormalities, regardless of their causal hol intake to ,25 g/day during the treat- treatment was discontinued in six patients relationship with the study drug. AEs ment period. (twoandfourintheplaceboand0.4mg/day were regarded as ADRs if the causal re- The fasting blood and urine samples pemafibrate groups, respectively). There- lationship could not be ruled out. were collected at least 10 h after the last fore, 160 patients completed the treatment meal. Apolipoproteins (Apos) were mea- Statistics for 24 weeks. Table 1 shows the character- sured through immunoassay. ApoB100 lev- Pemafibrate exposure in at least 100 pa- istics of the patients. No noteworthy dif- els were calculated by subtracting ApoB48 tients for 1 year was needed to satisfy the ferences were observed between the from ApoB. LDL-C, HDL-C, and remnant International Council for Harmonisation groups. Men accounted for 72.9% of the lipoprotein cholesterol (RemL-C) levels of Technical Requirements for Pharma- patients, the mean age of the participants were measured using the homogenous as- ceuticals for Human Use E1 guidelines was 60.5 years, and their mean BMI was says Determiner L LDL-C, MetaboLead for the evaluation of drug safety. Assum- 25.9 kg/m2. Approximately 60% of them HDL-C, and MetaboLead RemL-C (Kyowa ing that 10% of the patients would dis- had drinking habits, and two-thirds had Medex Co., Ltd., Tokyo, Japan), respec- continue participation in the study, 55 hypertension or fatty liver. The mean du- tively. Fibroblast growth factor 21 (FGF21) patients per group was the aim for en- ration of diabetes was 5.7 years, and was analyzed through an ELISA with Hu- rollment, considering that the power 44.6% of the patients were receiving man FGF21 ELISA (BioVendor, Brno, Czech of the primary efficacy analysis was one or two antidiabetic agents, with di- Republic). Common laboratory tests were .99%. peptidyl peptidase 4 inhibitor being the performed using standardized methods at The primary efficacy analysis was per- most frequently used drug (34.3%). Addi- LSI Medience Corporation, Tokyo, Japan. formedbasedonthefullanalysisset tionally, 39.2% of the patients received High-performance liquid chromatogra- through a last-observation-carried- statins (atorvastatin 23.1%; pitavastatin phy analyses (LipoSEARCH; Skylight Bio- forward (LOCF) method, imputing the 27.7%; rosuvastatin 35.4%; and other tech Inc., Akita, Japan) were performed last valid values to subsequent missing statins 13.8%). care.diabetesjournals.org Araki and Associates 541

Table 1—Patient characteristics Placebo (n = 57) Pemafibrate 0.2 mg/day (n = 54) Pemafibrate 0.4 mg/day (n =55) Age (years) 61.2 6 10.0 59.8 6 11.6 60.6 6 10.1 Age $65 years 35.1 (20) 35.2 (19) 34.5 (19) Men 66.7 (38) 79.6 (43) 72.7 (40) BMI (kg/m2)26.06 3.3 26.5 6 3.8 25.3 6 3.4 BMI $25 kg/m2 57.9 (33) 63.0 (34) 45.5 (25) Drinking habit 57.9 (33) 66.7 (36) 58.2 (32) Hypertension 64.9 (37) 59.3 (32) 60.0 (33) Fatty liver 56.1 (32) 59.3 (32) 47.3 (26) Duration of diabetes (years) 5.5 6 4.5 4.9 6 3.7 6.8 6 5.9 No antidiabetic drug 54.4 (31) 55.6 (30) 56.4 (31) One antidiabetic drug 22.8 (13) 25.9 (14) 23.6 (13) Sulfonylurea 3.7 (2) DPP-4 inhibitor 19.3 (11) 18.5 (10) 14.5 (8) a-Glucosidase inhibitor 5.5 (3) Glinide 3.5 (2) 1.9 (1) 1.8 (1) GLP-1 receptor agonist 1.9 (1) 1.8 (1) Two antidiabetic drugs 22.8 (13) 18.5 (10) 20.0 (11) Sulfonylurea/DPP-4 inhibitor 19.3 (11) 11.1 (6) 5.5 (3) Sulfonylurea/a-glucosidase inhibitor 3.6 (2) Sulfonylurea/GLP-1 receptor agonist 3.6 (2) DPP-4 inhibitor/a-glucosidase inhibitor 1.8 (1) 3.7 (2) 5.5 (3) DPP-4 inhibitor/glinide 1.8 (1) 1.8 (1) a-Glucosidase inhibitor/glinide 3.7 (2) Statin 40.4 (23) 33.3 (18) 43.6 (24) TG (mmol/L) 3.2 6 1.3 2.7 6 1.1 2.9 6 1.1 HDL-C (mmol/L) 1.2 6 0.3 1.2 6 0.3 1.3 6 0.7 HDL-C category ,1.0 (mmol/L) 28.1 (16) 25.9 (14) 20.0 (11) FPG (mmol/L) 7.7 6 1.1 7.7 6 1.1 7.4 6 1.1 Fasting insulin (pmol/L) 92.7 6 56.6 83.7 6 49.2 81.1 6 40.5 HOMA-IR 4.6 6 3.0 4.2 6 2.5 3.8 6 1.8

HbA1c (%) 7.0 6 0.5 6.9 6 0.4 7.0 6 0.4

HbA1c (mmol/mol) 52.9 6 5.0 52.4 6 4.7 52.6 6 4.7 Glycoalbumin (%) 16.9 6 2.5 17.1 6 2.0 17.1 6 2.2 eGFR (mL/min/1.73 m2)73.66 19.2 75.7 6 14.7 73.3 6 14.4 Data are presented as the mean 6 SD for continuous parameters and % (n) for categorical parameters. DPP-4, dipeptidyl peptidase 4; eGFR, estimated glomerular filtration rate; GLP-1, glucagon-like peptide 1.

Fasting serum TG levels in the placebo, imputation using the LOCF method. No showed significant reductions in non– 0.2 mg/day pemafibrate, and 0.4 mg/day sex differences were observed in the find- HDL-C, RemL-C, ApoB100, ApoB48, and pemafibrate groups decreased from ings. In each pemafibrate group, TG was ApoCIII levels and significant increases in 284.3 6 117.6 mg/dL (3.2 6 1.3 mmol/L), significantly reduced from week 4, and HDL-C and ApoA-I levels (Supplementary 240.3 6 93.5 mg/dL (2.7 6 1.1 mmol/L), the significance remained until week Fig. 3). LDL-C levels were not considerably and 260.4 6 95.9 mg/dL (2.9 6 1.1 24 (Fig. 1B)(P , 0.001 for each point). altered in these groups. As a result of the mmol/L), respectively, at baseline to The proportions of patients who achieved high-performance liquid chromatography 242.0 6 92.2 mg/dL (2.7 6 1.0 mmol/L), ,150 mg/dL (1.7 mmol/L) fasting se- analyses conducted at week 12, the cho- 129.0 6 71.5 mg/dL (1.5 6 0.8 mmol/L), rum TG levels at week 24 (LOCF) were lesterol content decreased in small and and 135.8 6 71.2 mg/dL (1.5 6 0.8 15.8%, 81.5%, and 70.9% in the pla- very small LDL, whereas it increased in mmol/L) at week 24 (LOCF). The percent- cebo, 0.2 mg/day pemafibrate, and large LDL in the pemafibrate groups age changes in fasting serum TG levels 0.4 mg/day pemafibrate groups, respec- (Supplementary Fig. 4). On the other at week 24 (LOCF) were 210.8% (P , tively. The distances to this target level hand, the cholesterol content increased 0.01), 244.3% (P , 0.001), and 245.1% at week 24 (LOCF) were 92.0 6 92.2 in medium, small, and very small HDL, (P , 0.001), respectively (Fig. 1A). More- mg/dL (1.0 6 1.0 mmol/L), 221.0 6 71.5 whereas it decreased in large HDL in these over, both of the pemafibrate groups had mg/dL (20.2 6 0.8 mmol/L), and 214.2 6 groups. statistically significant reductions in these 71.2 mg/dL (20.2 6 0.8 mmol/L), The changes in glycemic parame- levels compared with the placebo group respectively. ters were unclear (Fig. 2A–E). The 0.2 (P , 0.001, multiplicity adjusted). These With regard to other lipid-related mg/day pemafibrate group showed a findings were similar even without parameters, the pemafibrate groups significant decrease in HOMA-IR score 542 Pemafibrate and Diabetic Dyslipidemia Diabetes Care Volume 41, March 2018

as well as the incidence rates of AEs and ADRs, which were similar across the treat- ment groups (12,13). The unique quantitative and quali- tative lipid abnormalities frequently comorbid with type 2 diabetes based on IR involve the following mechanisms. First, impaired insulin action enhances adipocyte lipolysis by activating hormone- sensitive lipase, thereby liberating nonester- ified fatty acids (NEFAs) into the circulation. NEFA, which is taken up by the liver, is re- Figure 1—Percentage change in fasting serum TGs from baseline to week 24 (LOCF) (A), with values esterified to form TGs, thereby stimulating presented as the least squares mean 6 SEM estimated using ANCOVA with baseline level as the secretion of VLDLs (4). Hyperinsuline- a covariate, and fasting serum TG levels over time (B), with values presented as the mean 6 SD. mia secondary to IR is also suggested to , , ††† , **P 0.01, ***P 0.001 vs. baseline by ANCOVA. P 0.001 vs. placebo by ANCOVA with increase de novo lipogenesis through Dunnett test for multiplicity adjustment. Pema 0.2, pemafibrate 0.2 mg/day; Pema 0.4, pemafibrate 0.4 mg/day. augmenting the expression of carbohy- drate responsiveness element-binding protein and sterol regulatory element- compared with the placebo group. How- with acute kidney injury and liver func- binding protein-1c (4). Second, Niemann- ever, no significant changes in other gly- tion abnormality could not be ruled out. Pick C1-like 1 (NPC1L1) and microsomal TG cemic parameters were found between Abnormal elevations in levels of liver en- transfer protein mRNA expression was en- these groups. Both of the pemafibrate zymes, serum creatinine, estimated glo- hanced in the small intestine, leading to groups showed a slight increase from merular filtration rate, and CK in the the increases in chylomicron production the baseline in HbA1c level at week 24, pemafibrate groups were limited and and postprandial TG levels (14). Third, li- although this was not statistically signifi- comparable to those in the placebo poprotein lipase (LPL) activity was im- cant when compared with the placebo group. The liver enzyme levels decreased, paired, and the catabolism of increased group. The data were subjected to a and the renal function test results and CK chylomicrons and VLDLs was delayed, post hoc repeated-measures ANCOVA at levels were not significantly altered with leading to remnant lipoprotein accumula- weeks 4–24, in which the pemafibrate pemafibrate treatment (Supplementary tion, HDL-C reduction, and small, dense groups displayed significant reductions Table 1). LDL production (4). in fasting plasma glucose (FPG), fasting PPARa agonists have been implicated insulin, and HOMA-IR levels compared CONCLUSIONS to enhance hepatic NEFA uptake, NEFA with the placebo group (Fig. 2F–H). This study is the first to demonstrate the b-oxidation, and concomitant decreases FGF21 levels significantly increased in long-term efficacy and safety of pemafi- in de novo lipogenesis and VLDL produc- the pemafibrate groups (Supplementary brate treatment for over 24 weeks in pa- tion through which they modulate lipo- Fig. 5). tients with type 2 diabetes comorbid with protein profiles (15). Pemafibrate was The results of the meal tolerance test hypertriglyceridemia. Treatment with also shown to enhance b-oxidation– showed that the pemafibrate groups pemafibrate for 24 weeks remarkably re- related gene expression in human hepa- displayed a significantly reduced the duced the fasting serum TG levels by tocytes, murine hepatocytes, and rat area under the curve of 0–6.5 h for TG ;45%. The significant TG reduction was livers (16,17) and decrease de novo lipid at week 24 compared with the placebo stably maintained over 24 weeks. The synthesis (18), hepatic TG content (17), group, whereas the area under the curve proportion of patients who achieved and VLDL secretion in rat livers (18). of 0–6.5 h for plasma glucose and insulin ,150 mg/dL (1.7 mmol/L) TG levels at PPARa agonists are suggested to de- were not significantly altered (Sup- week 24 was .80% in the 0.2 mg/day crease intestinal cholesterol absorption, plementary Fig. 6). pemafibrate group but ;70% in the which may be mediated by decreases in The incidence rates of AEs and ADRs 0.4 mg/day pemafibrate group. This find- NPC1L1, microsomal TG transfer protein, were similar across the pemafibrate and ing may be attributed to the fact that the and ApoB mRNA (19,20). Pemafibrate placebo groups without statistically former group had lower baseline TG lev- also inhibited NPC1L1 mRNA expression significant differences (Table 2). Serious els by ;20 mg/dL (0.2 mmol/L) than the along with increased fecal cholesterol AEs were observed in three (5.3%), three latter group. Not only TG but also other excretion in LDL receptor knockout mice (21) (5.6%), and two (3.6%) patients, respec- markers of TG-rich lipoproteins were and inhibited ApoB mRNA expression tively, in the placebo, 0.2 mg/day pemafi- dramatically ameliorated. Additionally, in apoE2/E2 knockin mice (22). In the cur- brate, and 0.4 mg/day pemafibrate HDL-C and ApoA-I levels increased. These rent study, the levels of ApoB48, a major groups, and the causal relationship with changes were accompanied by favorable component of chylomicrons, were signif- the study treatment was ruled out for all shifting in the LDL and HDL atherogenic icantly reduced with pemafibrate treat- groups. AEs leading to discontinuation of profiles by subclasses. These comprehen- ment, which presumably reflects the participation in the study were observed sive effects on lipoprotein profiles were reduction in intestine-derived chylomi- in four patients in the 0.4 mg/day pema- similar to those observed in the previous cron production and/or its stimulated ca- fibrate group, and a causal relationship studies including patients without diabetes tabolism. Furthermore, PPARa agonists care.diabetesjournals.org Araki and Associates 543

with pemafibrate also, considering that very large and large HDL-C levels de- creased in the 0.4 mg/day pemafibrate group more than in the 0.2 mg/day pe- mafibrate group, whereas medium, small, and very small HDL-C levels increased in both groups in the current study. The same trend has been observed in previ- ous studies (12,13). The effects of pemafibrate on cardio- vascular outcomes are being assessed in a global large-scale trial in patients with type 2 diabetes comorbid with atherogenic dyslipidemia (PROMINENT; clinical trial reg. no. NCT03071692) (26). The biphasic dose response, which was not clear in fenofi- brate treatment to the best of our knowl- edge, may be related to the fact that LY518674 and pemafibrate are both more potent and selective PPARa agonists than fenofibrate (16,27). Pemafibrate is primarily eliminated via the liver, whereas fenofibrate and gemfibrozil are eliminated via kidneys (26,28). The above-mentioned differences may be, at least in part, asso- ciated with different responses in several clinical laboratory tests. In the present and previous studies of pemafibrate compared with placebo and/or fenofibrate (12,13,29), the effects of pemafibrate on serum creatinine, ALT, g-glutamyl transferase, and homocys- teine levels appeared comparable to that of placebo or smaller than that of Figure 2—Change from baseline to week 24 (LOCF) in FPG (A), fasting insulin (B), HOMA-IR (C), fenofibrate. Rather, pemafibrate even 6 glycoalbumin (D), and HbA1c (E), with values presented as the mean SD, and the change from decreased the liver enzyme levels and baseline to weeks 4–24 in FPG (F), fasting insulin (G), and HOMA-IR (H) estimated by post hoc – 6 affected the serum creatinine levels to a repeated-measures ANCOVA for weeks 4 24, with values presented as the least squares mean fi SEM. *P , 0.05, **P , 0.01 vs. baseline by one-sample t test. †P , 0.05 vs. placebo by ANCOVA with lesser extent than feno brate. Further- baseline level as a covariate. ‡P , 0.05, ‡‡P , 0.01 vs. placebo by repeated-measures ANCOVA. more, these effects of pemafibrate mono- therapy were similar to pemafibrate and statin combination therapy (13). Gemfi- brozil shares similar adverse effects on have been shown to increase the expres- observed in a dose-response study of an- the clinical laboratory tests and is associ- sion of LPL and inhibit that of ApoCIII, other PPARa agonist, LY518674, compar- ated with a relatively higher risk of rhab- which inhibits LPL activity (15). Pemafi- ing placebo and 200 mg/day fenofibrate domyolysis, which is even increased with brate suppressed ApoC3 mRNA expres- (24). The baseline lipid profiles were sim- statin combination therapy (28,30). In the sion and enhanced postheparin plasma ilar in both studies, although only 14.6% study of LY518674 (24), the incidence LPL activity in LDL receptor knockout of patients had diabetes in the latter rates of serum creatinine levels greater mice (21). In fact, pemafibrate remark- study. The potential biphasic dose re- than the ULN and ALT levels .1.53 ably decreased the ApoCIII levels in the sponse of HDL-C increase and non–HDL-C ULN were over three times higher in the present and previous clinical studies. reduction in the current study was simi- 25–100 mg/day LY518674 and 200 mg/day Through these mechanisms (i.e., sup- larlyobservedinthestudyofLY518674 fenofibrate groups than in the placebo pressing the production and enhancing (24). Such a dose response of HDL-C in- group, whereas both of them in the the catabolism of TG-rich lipoproteins), crease with LY518674 may be attributed pemafibrate groups were comparable or pemafibrate was suggested to increase to enhanced ApoA-I turnover, which was lower than those in the placebo group levels of HDL-C and decrease levels of suggested by increased production and from post hoc analyses in the current small, dense LDL, which has even higher fractional catabolic rate of ApoA-I(25) study using similar cutoff levels (Supple- atherogenicity (23). and may be, at least in part, reflected mentary Table 2). The differences in the The overall results of lipid profiles in to that of non–HDL-C reduction. ApoA-I safety profiles supported a good risk/ the current study were similar to those turnover may be similarly enhanced benefit balance of pemafibrate. 544 Pemafibrate and Diabetic Dyslipidemia Diabetes Care Volume 41, March 2018

fi Table 2—Summary of AEs and ADRs type 2 diabetes. The good risk/bene tbal- fi fi Placebo Pemafibrate 0.2 mg/day Pemafibrate 0.4 mg/day ance of pema brate was con rmed in this (n =57) (n = 54) (n =55) population, which was similar to that in previous studies in patients with hypertri- Total AEs 41 (71.9) 36 (66.7) 33 (60.0) vs. placebo P =0.681 P =0.232 glyceridemia with or without diabetes, over a long period of 24 weeks. These find- Serious AEs 3 (5.3) 3 (5.6) 2 (3.6) ings provide significant information on the AEs leading to discontinuation 0 0 4 (7.3) management of lipid abnormalities in pa- Total ADRs 7 (12.3) 6 (11.1) 9 (16.4) tients with type 2 diabetes comorbid with vs. placebo P =1.000 P =0.597 hypertriglyceridemia. Serious ADRs 0 0 0 ADRs leading to discontinuation 0 0 2 (3.6) Acknowledgments. We acknowledge the in- Laboratory tests vestigators and patients who participated in this AST .33 ULN 0 0 0 study (see APPENDIX). . 3 AST 5 ULN 0 0 0 Funding. This work was supported by Kowa ALT .33 ULN 1 (1.8) 0 0 Company, Ltd. ALT .53 ULN 0 0 0 The study sponsor had a role in the study de- GGT .33 ULN 4 (7.0) 2 (3.7) 0 sign, data collection, data analysis, data inter- GGT .53 ULN 4 (7.0) 0 0 pretation, and writing of the report. CK .43 ULN 0 1 (1.9) 0 Duality of Interest. E.A. reports receiving per- CK .53 ULN 0 1 (1.9) 0 sonal fees from Kowa during the conduct of the CK .103 ULN 0 0 0 study as well as grants from Nippon Boehringer Serum Ingelheim, Novo Nordisk Pharma, Ono Pharma- creatinine .1.5 mg/dL 2 (3.5) 1 (1.9) 2 (3.6) ceutical, Sanofi,DaiichiSankyo,MitsubishiTanabe Serum Pharma, Novartis Pharma, Kowa Pharmaceutical, creatinine .2.0 mg/dL 0 1 (1.9) 1 (1.8) Astellas Pharma, AstraZeneca, Takeda Pharmaceu- tical, Taisho Toyama Pharmaceutical, and Pfizer Data are presented as the number of patients (%). GGT, g-glutamyl transferase. P values were Japan and personal fees from MSD, Nippon estimated by Fisher exact test. Boehringer Ingelheim, Novo Nordisk Pharma, Ono Pharmaceutical, Sanofi, Daiichi Sankyo, Mitsubishi Tanabe Pharma, Novartis Pharma, Kowa Pharma- ceutical, Astellas Pharma, AstraZeneca, Takeda In terms of glycemic parameters, the vascular events. The effects of PPARa Pharmaceutical, Taisho Toyama Pharmaceutical, post hoc repeated-measures ANCOVA agonists on cardiovascular events in pa- and Eli Lilly Japan outside the submitted work. showed that pemafibrate decreased the tients with type 2 diabetes were exam- S.Y. reports receiving grants and personal fees from fasting glucose, insulin, and HOMA-IR lev- ined in the FIELD and ACCORD lipid Kowa during the conduct of the study as well as grants from Nippon Boehringer Ingelheim, Otsuka els, although the prespecified analyses studies (9,40). These studies demon- Pharmaceutical, Shionogi, Bayer Yakuhin National In- did not show a clear trend. The results strated that cardiovascular events were stitute of Biomedical Innovation, MSD, Japan To- of previous studies on the effects of significantly suppressed in the subgroup bacco, Kyowa Medex, Takeda Pharmaceutical, Sanwa PPARa agonists other than pemafibrate of patients with high TG and low HDL-C Kagaku Kenkyusho, Astellas Pharma, Daiichi Sankyo, – fi Mochida Pharmaceutical, AstraZeneca, Izumisano on IR were inconsistent (31 35). Pema - levels and suggested that diabetic micro- City, Kaizuka City, Hayashibara, Teijin Pharma, Kaken brate has been suggested to improve IR in angiopathy might be prevented. Fur- Pharmaceutical, and Kissei Pharmceutical and personal the hyperinsulinemic-euglycemic–clamp ther large-scale studies on the effect of fees from Otsuka Pharmaceutical, Shionogi, Bayer study (36) and pooled analyses of previ- pemafibrate on diabetic complications Yakuhin, MSD, Takeda Pharmaceutical, Sanwa ous studies (37). Moreover, markedly el- and cardiovascular outcomes are neces- Kagaku Kenkyusho, Ono Pharmaceutical, Astellas Pharma, Daiichi Sankyo, AstraZeneca, Medical Re- evated FGF21 levels and reduced ApoCIII sary. Second, all patients were Japanese, view, Skylight Biotech, Kaken Pharmaceutical, levels with pemafibrate treatment may many patients had relatively mild type 2 Pfizer Japan, Bristol-Myers Squibb, Amgen Astellas positively impact the reduction of IR be- diabetes, many antidiabetic agents were BioPharma, Sanofi, and Toa Eiyo outside the sub- cause an FGF21 analog ameliorated IR prohibited, and changes in the class and mitted work. In addition, S.Y. has a Fujirebio patent and glucose metabolism (38) and an dosage regimen of antidiabetic agents pending. H.A. reports receiving personal fees from Kowa during the conduct of the study as well as ApoCIII antisense improved insulin sensi- were restricted even for nonprohibited grants from Daiichi Sankyo and personal fees from tivity (39). The changes in fasting glycemic drugs. Therefore, further investigation is Daiichi Sankyo, MSD, and Otsuka Pharmaceutical parameters were inconsistent with those needed to clarify whether the findings of outside the submitted work. K.Y. reports per- sonal fees from Kowa during the conduct of the in HbA1c and glycoalbumin levels, which the current study can be generalized to reflects the mean plasma glucose levels other races or patients with more severe study as well as grants from Astellas Pharma, Otsuka Pharmaceutical, Daiichi Sankyo, Takeda over the past 1–2 months and 2 weeks, diabetes. Pharmaceutical, Mitsubishi Tanabe Pharma, and respectively. Therefore, further investiga- Bristol-Myers Squibb and personal fees from Astellas tion is needed to confirm the effects of Conclusion Pharma,AstraZeneca,Eisai,MSD,OnoPharmaceuti- pemafibrate on glucose metabolism. Pemafibrate, a novel selective PPARa mod- cal, Kyowa Hakko Kirin, Kowa Pharmaceutical, fi Shionogi, Daiichi Sankyo, Takeda Pharmaceuti- The current study has the following ulator, demonstrated excellent ef cacy cal, Mitsubishi Tanabe Pharma, Pfizer Japan, limitations. First, it was not designed to in the amelioration of lipid abnormalities and Mochida Pharmaceutical outside the submit- investigate the effects of pemafibrate on and was well tolerated in patients with ted work. J.S. reports receiving personal fees from care.diabetesjournals.org Araki and Associates 545

Kowa during the conduct of the study as well as to the interpretation of the data and substan- General Practice Research Database. Diabetologia personal fees from Astellas Pharma, AstraZeneca, tially contributed to writing and critically review- 2008;51:1639–1645 Nippon Boehringer Ingelheim, Eli Lilly Japan, Kyowa ing the manuscript. S.Y., H.A., K.Y., J.S., T.I., J.N., 4. Verges` B. Pathophysiology of diabetic dyslipi- Hakko Kirin, Mitsubishi Tanabe Pharma, MSD, Novo H.M., N.Y., Y.T., H.W., and S.I. contributed to the daemia: where are we? Diabetologia 2015;58: Nordisk Pharma, Ono Pharmaceutical, Pfizer Japan, concept, design, and execution of the study; the 886–899 Sanofi, Sanwa Kagaku Kenkyusho, Sumitomo interpretation of the data; and critical review of 5. Colhoun HM, Betteridge DJ, Durrington PN, Dainippon Pharma, Taisho Toyama Pharmaceutical, the manuscript. H.S. contributed to the concept, et al.; CARDS investigators. Primary prevention of and Takeda Pharmaceutical outside the submitted design, and execution of the study and to the cardiovascular disease with atorvastatin in type 2 work. T.I. reports receiving personal fees from Kowa interpretation of the data. H.S. is the guarantor diabetes in the Collaborative Atorvastatin Diabetes during the conduct of the study as well as grants of this work and, as such, had full access to all the Study (CARDS): multicentre randomised placebo- from Nippon Boehringer Ingelheim, Mitsubishi data in the study and takes responsibility for the controlled trial. Lancet 2004;364:685–696 Tanabe Pharma, Novo Nordisk Pharma, Sanofi, integrity of the data and the accuracy of the data 6. Kearney PM, Blackwell L, Collins R, et al.; Cho- Astellas Pharma, Kowa Pharmaceutical, AstraZeneca, analysis. lesterol Treatment Trialists’ (CTT) Collabora- Ono Pharmaceutical, Eli Lilly Japan, Sumitomo Prior Presentation. Parts of this study were tors. Efficacy of cholesterol-lowering therapy in Dainippon Pharma, and MSD and personal fees presented in abstract form at The 59th Annual 18,686 people with diabetes in 14 randomised from Nippon Boehringer Ingelheim, Mitsubishi Ta- Meeting of the Japan Diabetes Society, Kyoto, trials of statins: a meta-analysis. Lancet 2008; nabe Pharma, Novo Nordisk Pharma, Taisho Toyama Japan, 19–21 May 2016, and the 52nd Annual 371:117–125 Pharmaceutical, Sanofi, Astellas Pharma, Kowa Meeting of the European Association for the 7. Sone H, Tanaka S, Tanaka S, et al.; Japan Di- Pharmaceutical, Novartis Pharma, AstraZeneca, Study of Diabetes, Munich, Germany, 12–16 Sep- abetes Complications Study Group. Serum level of Ono Pharmaceutical, Takeda Pharmaceutical, Eli tember 2016. triglycerides is a potent risk factor comparable to Lilly, Kissei Pharmaceutical, Sumitomo Dainippon LDL cholesterol for coronary heart disease in Jap- Pharma, and MSD outside the submitted work. J.N. anese patients with type 2 diabetes: subanalysis reports receiving personal fees from Kowa during the Appendix of the Japan Diabetes Complications Study (JDCS). conduct of the study. H.M. reports receiving personal The study sites and their principal investigators J Clin Endocrinol Metab 2011;96:3448–3456 fees from Kowa during the conduct of the study. N.Y. are as follows: NTT East Sapporo Hospital (So 8. Turner RC, Millns H, Neil HA, et al. Risk factors reports receiving personal fees from Kowa during Nagai), Iwate Medical University Hospital for coronary artery disease in non-insulin dependent the conduct of the study as well as personal fees (Noriko Takebe), Juntendo University Hospi- diabetes mellitus: United Kingdom Prospective Di- from Novo Nordisk Pharma, Sanofi, and MSD out- tal (Fuki Ikeda), Aichi Medical University Hospital abetes Study (UKPDS: 23). BMJ 1998;316:823–828 side the submitted work. Y.T. reports receiving per- (J.N.), Shiga University of Medical Science 9. Ginsberg HN, Elam MB, Lovato LC, et al.; sonal fees from Kowa during the conduct of the study Hospital (H.M.), Yamaguchi University Hospital ACCORD Study Group. Effects of combination lipid (Yasuharu Ohta), Kyushu University Hospital as well as grants from Astellas Pharma, MSD, Ono therapy in type 2 diabetes mellitus. N Engl J Med (T.I.), Kumamoto University Hospital (Takeshi Pharmaceutical, Taisho Toyama Pharmaceutical, 2010;362:1563–1574 Sumitomo Dainippon Pharma, Takeda Pharmaceuti- Matsumura), Chiba University Hospital (Minoru 10. Scott R, O’Brien R, Fulcher G, et al.; Fenofi- cal, Nippon Boehringer Ingerheim, Kyowa Hakko Takemoto), Sugiura Clinic (Toshiyuki Sugiura), brate Intervention and Event Lowering in Diabetes Kirin, Sanwa Kagaku Kenkyusho, Shionogi, Daiichi Musashi Fujisawa Central Clinic (Seiki Wada), (FIELD) Study Investigators. Effects of fenofibrate Sankyo, Pfizer Japan, Bristol-Myers Squibb, and Eli Nakayama Clinic (Mikihiro Nakayama), Goshi treatment on cardiovascular disease risk in 9,795 Lilly Japan and personal fees from Astellas Pharma, Hospital (Takeo Naito), Tao Internal Medicine individuals with type 2 diabetes and various com- AstraZeneca, MSD, Ono Pharmaceutical, Kissei Phar- Clinic (Tsuyoshi Tao), BOOCS Clinic Fukuoka ponents of the metabolic syndrome: the Fenofibrate maceutical, Sanofi, Taisho Toyama Pharmaceutical, (Kazuyuki Saito), Diabetes Center, Shin-Koga Intervention and Event Lowering in Diabetes (FIELD) Sumitomo Dainippon Pharma, Takeda Pharma, Hospital (Shoichi Akazawa and Eiji Kawasaki), study. Diabetes Care 2009;32:493–498 Mitsubishi Tanabe Pharma, Boehringer Ingerheim Kurihara Diabetic Care Clinic (Yoshio Kurihara), 11. Rubins HB, Robins SJ, Collins D, et al. Diabetes, Japan, and Novo Nordisk Pharmaceutical outside Okuguchi Clinic of Internal Medicine (Fuminobu plasma insulin, and cardiovascular disease: sub- the submitted work. H.W. reports receiving per- Okuguchi), Naka Kinen Clinic (Takeshi Osonoi), group analysis from the Department of Veterans sonal fees from Kowa during the conduct of the Tomonaga Clinic (Osamu Tomonaga), Kurashiki Affairs high-density lipoprotein intervention trial study as well as grants from Astellas Pharma, Sanofi, Central Hospital (Takashi Matsuoka), Kanauchi (VA-HIT). Arch Intern Med 2002;162:2597–2604 Mitsubishi Tanabe Pharma, Novo Nordisk Pharma, Medical Clinic (Rie Wada), Osaka Gyoumeikan 12. Ishibashi S, Yamashita S, Arai H, et al.; K-877- AstraZeneca, Takeda Pharmaceutical, Novartis Hospital (Shinya Makino), Osaka Ekisaikai Hos- 04 Study Group. Effects of K-877, a novel selective Pharma, Nippon Boehringer Ingelheim, MSD, pital (Haruyuki Taguchi), Tohoku Medical and PPARa modulator (SPPARMa), in dyslipidaemic Sumitomo Dainippon Pharma, Eli Lilly Japan, Pharmaceutical University Wakabayashi Hospi- patients: a randomized, double blind, active- Ono Pharmaceutical, Kyowa Hakko Kirin, Daiichi tal, formerly NTT East Tohoku Hospital (Kazumi and placebo-controlled, phase 2 trial. Atheroscle- Sankyo, Terumo, Pfizer Japan, BenefitoneHealth Yamato), Tokyo-Eki Center-Building Clinic (Arihiro rosis 2016;249:36–43 care, Mochida Pharmaceutical, Nitto Boseki, Taisho Kiyosue), Sakakibara Sapia Tower Clinic (Makiko 13. Arai H, Yamashita S, Yokote K, Araki E, Toyama Pharmaceutical, Johnson & Johnson, and Abe), Chikamori Hospital (Yoshitaka Kumon), Suganami H, Ishibashi S; K-877 Study Group. Effi- Kowa and personal fees from Astellas Pharma, Ayame Medical Clinic (Hideo Ayame), Ehime Rosai cacy and safety of K-877, a novel selective peroxi- Sanofi, Mitsubishi Tanabe Pharma, Novo Nordisk Hospital (Kazuaki Nakai), Saiseikai Matsuyama some proliferator-activated receptor a modulator Pharma, Kowa Pharmaceutical, AstraZeneca, Hospital (Hiroaki Miyaoka), Matsuyama Red Cross (SPPARMa), in combination with statin treatment: Takeda Pharmaceutical, Novartis Pharma, Nippon Hospital (Shiori Kondo), Matsuba Clinic (Ikuro two randomised, double-blind, placebo-controlled Boehringer Ingelheim, MSD, Sumitomo Dainippon Matsuba), and Manda Memorial Hospital (Shinji clinical trials in patients with dyslipidaemia. Athero- Pharma, Eli Lilly Japan, Sanwa Kagaku Kenkyusho, Taneda). sclerosis 2017;261:144–152 Ono Pharmaceutical, Kissei Pharmaceutical, and 14. Lally S, Tan CY, Owens D, Tomkin GH. Mes- FUJIFILM Pharma outside the submitted work. References senger RNA levels of genes involved in dysregu- H.S. is an employee of Kowa. S.I. reports receiving 1. Tancredi M, Rosengren A, Svensson AM, et al. lation of postprandial lipoproteins in type 2 personal fees from Kowa during the conduct of the Excess mortality among persons with type 2 di- diabetes: the role of Niemann-Pick C1-like 1, study as well as grants from Astellas Pharma, Daiichi abetes. N Engl J Med 2015;373:1720–1732 ATP-binding cassette, transporters G5 and G8, Sankyo,TeijinPharma,TakedaPharmaceutical, 2. Haffner SM, Lehto S, Ronnemaa¨ T, Pyor¨ al¨ aK,¨ and of microsomal triglyceride transfer protein. Ono Pharmaceutical, Taisho Toyama Pharmaceutical, Laakso M. Mortality from coronary heart disease Diabetologia 2006;49:1008–1016 and Nippon Boehringer Ingelheim and personal fees in subjects with type 2 diabetes and in nondia- 15. Staels B, Dallongeville J, Auwerx J, Schoonjans from MSD, AstraZeneca, Sanwa Kagaku Kenkyusho, betic subjects with and without prior myocardial K, Leitersdorf E, Fruchart JC. Mechanism of action and Nippon Boehringer Ingelheim outside the infarction. N Engl J Med 1998;339:229–234 of fibrates on lipid and lipoprotein metabolism. submitted work. 3. Mulnier HE, Seaman HE, Raleigh VS, et al. Risk Circulation 1998;98:2088–2093 Author Contributions. E.A. contributed to the of myocardial infarction in men and women with 16. Raza-Iqbal S, Tanaka T, Anai M, et al. Tran- concept, design, and execution of the study and type 2 diabetes in the UK: a cohort study using the scriptome analysis of K-877 (a novel selective 546 Pemafibrate and Diabetic Dyslipidemia Diabetes Care Volume 41, March 2018

PPARa modulator (SPPARMa))-regulated genes subjects with the metabolic syndrome. Arterioscler 32. Cariou B, Hanf R, Lambert-Porcheron S, et al. in primary human hepatocytes and the mouse Thromb Vasc Biol 2009;29:140–146 Dual peroxisome proliferator-activated receptor liver. J Atheroscler Thromb 2015;22:754–772 26. Blair HA. Pemafibrate: first global approval. a/d agonist GFT505 improves hepatic and periph- 17. Takei K, Han SI, Murayama Y, et al. The selec- Drugs 2017;77:1805–1810 eral insulin sensitivity in abdominally obese sub- tive PPARalpha modulator K-877 efficiently acti- 27. Singh JP, Kauffman R, Bensch W, et al. jects. Diabetes Care 2013;36:2923–2930 vates the PPARalpha pathway and improves lipid Identification of a novel selective peroxisome 33. Black RN, Ennis CN, Young IS, Hunter SJ, metabolism in mice. J Diabetes Investig 2017;8: proliferator-activated receptor alpha agonist, Atkinson AB, Bell PM. The peroxisome proliferator- 446–452 2-methyl-2-(4-3-[1-(4-methylbenzyl)-5-oxo-4,5- activated receptor alpha agonist fenofibrate has 18. Takizawa T, Inokuchi Y, Goto S, et al. The dihydro-1H-1,2,4-triazol-3-yl]propylphenoxy)- no effect on insulin sensitivity compared to ator- mechanism of K-877, a highly potent and selective propanoic acid (LY518674), that produces marked vastatin in type 2 diabetes mellitus; a randomised, pparalpha modulator, on regulation of synthesis, changes in serum lipids and apolipoprotein A-1 double-blind controlled trial. J Diabetes Complica- secretion and metabolism of triglycerides and expression. Mol Pharmacol 2005;68:763–768 tions 2014;28:323–327 cholesterol (Abstract). Circulation 2013;128: 28. Wiggins BS, Saseen JJ, Page RL 2nd, et al,; 34. Shiochi H, Ohkura T, Fujioka Y, et al. Bezafi- A12867 American Heart Association Clinical Pharmacology brate improves insulin resistance evaluated using 19. Jia L, Betters JL, Yu L. Niemann-pick C1-like 1 Committee of the Council on Clinical Cardiology; the glucose clamp technique in patients with (NPC1L1) protein in intestinal and hepatic choles- Council on Hypertension; Council on Quality of type 2 diabetes mellitus: a small-scale clinical terol transport. Annu Rev Physiol 2011;73:239– Care and Outcomes Research; and Council on Func- study. Diabetol Metab Syndr 2014;6:113 259 tional Genomics and Translational Biology. Recom- 35. Taniguchi A, Fukushima M, Sakai M, et al. 20. Sandoval JC, Nakagawa-Toyama Y, Masuda D, mendations for management of clinically significant Effects of bezafibrate on insulin sensitivity and et al. Fenofibrate reduces postprandial hypertri- drug-drug interactions with statins and select insulin secretion in non-obese Japanese type 2 glyceridemia in CD36 knockout mice. J Atheros- agents used in patients with cardiovascular disease: diabetic patients. Metabolism 2001;50:477–480 cler Thromb 2010;17:610–618 ascientific statement from the American Heart As- 36. Matsuba I, Matsuba R, Ishibashi S, et al. The 21. Takei K, Nakagawa Y, Wang Y, et al. Effects of sociation. Circulation 2016;134:e468–e495 effects of selective PPAR alpha modulator, K-877, K-877, a novel selective PPARa modulator, on 29. Ishibashi S, Arai H, Yokote K, Araki E, on the insulin sensitivity evaluated by glucose small intestine contribute to the amelioration of Suganami H, Yamashita S. Efficacy and safety of clamp test (Abstract). J-Stage 2016;59:S462 [in hyperlipidemia in low-density lipoprotein receptor pemafibrate (K-877), a selective peroxisome Japanese] knockout mice. J Pharmacol Sci 2017;133:214–222 proliferator-activated receptor a modulator 37. Araki E, Ishibashi S, Yamashita S, et al. A highly 22. Hennuyer N, Duplan I, Paquet C, et al. The (SPPARMa), in patients with dyslipidemia: results potent and specific PPAR alpha agonist, K-877, novel selective PPARa modulator (SPPARMa) from a 24-week, randomized, double blind, improves lipid profiles and insulin sensitivity in pemafibrate improves dyslipidemia, enhances active-controlled, phase 3 trial. J Clin Lipidol. 27 dyslipidaemia subjects; an integrated analysis of reverse cholesterol transport and decreases in- October 2017 [Epub ahead of print]. https://doi 3 phase 2/3 trials (Abstract). Diabetologia 2014; flammation and atherosclerosis. Atherosclerosis .org/10.1016/j.jacl.2017.10.006 57:S272 2016;249:200–208 30. Catapano AL, Graham I, De Backer G, et al.; 38. Gaich G, Chien JY, Fu H, et al. The effects of 23. Hoogeveen RC, Gaubatz JW, Sun W, et al. Authors/Task Force Members. 2016 ESC/EAS Guide- LY2405319, an FGF21 analog, in obese human Small dense low-density lipoprotein-cholesterol lines for the Management of Dyslipidaemias: the subjects with type 2 diabetes. Cell Metab 2013; concentrations predict risk for coronary heart dis- Task Force for the Management of Dyslipidaemias 18:333–340 ease: the Atherosclerosis Risk In Communities of the European Society of Cardiology (ESC) and 39. Digenio A, Dunbar RL, Alexander VJ, et al. (ARIC) study. Arterioscler Thromb Vasc Biol European Atherosclerosis Society (EAS) Developed Antisense-mediated lowering of plasma apolipo- 2014;34:1069–1077 with the special contribution of the European Asso- protein C-III by volanesorsen improves dyslipide- 24. Nissen SE, Nicholls SJ, Wolski K, et al. Effects ciation for Cardiovascular Prevention & Rehabilita- mia and insulin sensitivity in type 2 diabetes. of a potent and selective PPAR-alpha agonist in tion (EACPR). Atherosclerosis 2016;253:281–344 Diabetes Care 2016;39:1408–1415 patients with atherogenic dyslipidemia or hyper- 31. Bajaj M, Suraamornkul S, Hardies LJ, Glass L, 40. Keech A, Simes RJ, Barter P, et al.; FIELD study cholesterolemia: two randomized controlled tri- Musi N, DeFronzo RA. Effects of peroxisome pro- investigators. Effects of long-term fenofibrate als. JAMA 2007;297:1362–1373 liferator-activated receptor (PPAR)-alpha and therapy on cardiovascular events in 9795 people 25. Millar JS, Duffy D, Gadi R, et al. Potent and PPAR-gamma agonists on glucose and lipid me- with type 2 diabetes mellitus (the FIELD study): selective PPAR-alpha agonist LY518674 upregulates tabolism in patients with type 2 diabetes mellitus. randomised controlled trial. Lancet 2005;366: both ApoA-I production and catabolism in human Diabetologia 2007;50:1723–1731 1849–1861