Ida et al. Cardiovasc Diabetol (2019) 18:38 https://doi.org/10.1186/s12933-019-0845-x Cardiovascular Diabetology

ORIGINAL INVESTIGATION Open Access Efcacy and safety of pemafbrate administration in patients with dyslipidemia: a systematic review and meta‑analysis Satoshi Ida*, Ryutaro Kaneko and Kazuya Murata

Abstract Background: Using a meta-analysis of randomized controlled trials (RCTs), this study aimed to investigate the ef- cacy and safety of pemafbrate, a novel selective peroxisome proliferator-activated receptor α modulator, in patients with dyslipidemia. Methods: A search was performed using the MEDLINE, Cochrane Controlled Trials Registry, and ClinicalTrials.gov databases. We decided to employ RCTs to evaluate the efects of pemafbrate on lipid and glucose metabolism- related parameters in patients with dyslipidemia. For statistical analysis, standardized mean diference (SMD) or odds ratio (OR) and 95% confdence intervals (CIs) were calculated using the random efect model. Results: Our search yielded seven RCTs (with a total of 1623 patients) that satisfed the eligibility criteria of this study; hence, those studies were incorporated into this meta-analysis. The triglyceride concentration signifcantly decreased in the pemafbrate group (SMD, 1.38; 95% CI, 1.63 to 1.12; P < 0.001) than in the placebo group, with a reduc- tion efect similar to that exhibited− by fenofbrate.− Compared− with the placebo group, the pemafbrate group also showed improvements in high-density and non-high-density lipoprotein cholesterol levels as well as in homeostasis model assessment for insulin resistance. Furthermore, the pemafbrate group showed a signifcant decrease in hepa- tobiliary enzyme activity compared with the placebo and fenofbrate groups; and, total adverse events (AEs) were signifcantly lower in the pemafbrate group than in the fenofbrate group (OR, 0.60; 95% CI, 0.49–0.73; P < 0.001). In contrast, the low-density lipoprotein cholesterol level was signifcantly higher in the pemafbrate group than in the placebo (P 0.006) and fenofbrate (P < 0.001) groups. = Conclusions: The lipid profle signifcantly improved in the pemafbrate group than in the placebo group. In addition to the pemafbrate group having an improved lipid profle, which was comparable with that of the fenofbrate group, the AEs were signifcantly lower than in the fenofbrate group and an improvement in hepatobiliary enzyme activity was also recognized. However, we believe that actual clinical data as well as long-term efcacy and safety need to be investigated in the future. Keywords: Fibrate, Meta-analysis, Dyslipidemia, Pemafbrate, Adverse events, Lipid profle, Hepatobiliary enzyme activity

*Correspondence: [email protected] Department of Diabetes and Metabolism, Ise Red Cross Hospital, 1‑471‑2, Funae, 1‑Chome, Ise‑shi, Mie 516‑8512, Japan

© The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat​iveco​mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat​iveco​mmons​.org/ publi​cdoma​in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 2 of 14

Background Trials Registry (from 1960), and ClinicalTrials.gov. Previous studies have demonstrated that lipid control is Te search strategy included terms of “(pemafbrate or important for preventing cardiovascular disease onset [1, selective peroxisome proliferator-activated receptor or 2]. In particular, the fnding that cardiovascular disease selective peroxisome proliferator-activated receptor or onset can be prevented by reducing low-density lipo- selective modulator of peroxisome proliferator-activated protein cholesterol (LDL-C) levels via statin administra- receptor or selective modulator of peroxisome prolif- tion [1] has achieved high clinical priority. However, the erator-activated receptor or selective PPAR or selective problem here is that the beneft of statin administration modulator of PPAR) AND (Randomized Controlled Trial to cardiovascular disease inhibition is only approximately or Controlled Clinical Trial or Randomized or Rand- 20%, the remaining 80% being residual risk [3]. Target omized or placebo or randomly or assigned).” We decided lipids for arteriosclerosis prevention, including triglycer- to include RCTs to evaluate the efect of pemafbrate ides (TG) and high-density lipoprotein cholesterol (HDL- on lipid and glucose metabolism-related parameters in C) except LDL-C, and hypertriglyceridemia and low patients with dyslipidemia. Te RCTs included compared HDL-C-emia are reportedly associated with cardiovas- pemafbrate with placebo and lipid-lowering drugs irre- cular disease onset [4]. As a drug against hyperTG-emia spective of the presence of diet/exercise therapy or the and low HDL-C-emia, a fbrate-system drug capable use of lipid-lowering drugs. Exclusion criteria included of strong TG-lowering and HDL-C-elevating actions is studies that were not RCT, research on animals, research being clinically used to activate peroxisome proliferator- without sufcient data to perform analysis, and cases activated receptor α (PPARα) [5]. Some meta-analyses of duplicate literature. In cases of difculty in interpre- have reported the possibility of the suppression of cardio- tation, the senior author decided to consult with other vascular disease development by fbrate administration reviewers (RK and KM). [6, 7]; this may be useful in remedying the residual risk of arteriosclerosis that cannot be sufciently controlled Data extraction and quality assessment by statin administration alone. However, manifestation We created a data extraction form describing the char- of side efects, such as hepatic disorder, kidney disorder, acteristics of the research included in each study (key or rhabdomyolysis, may be problematic for fbrate-type author’s name, publication year, study location, sam- drugs and may occasionally make it difcult to continue ple size, patient’s baseline information, basic treatment, administration [8, 9]. Tus, fbrate formulations are pres- and treatment duration). Continuous variables were ently considered problematic for wide-scale clinical use. expressed as the mean value, standard deviation, stand- Pemafbrate reportedly reduces TG, which is consid- ard error or 95% confdence intervals (CIs), and binary ered as a risk factor for the development of cardiovas- variables were expressed in terms of percentage (%). In cular diseases, or increases HDL-C levels. Furthermore, case of a study comparing one placebo group and two or pemafbrate is a PPARα modulator (SPPARMα) with more intervention groups, we treated it as two or more extremely high selectivity to PPARs subtypes, devel- studies sharing the placebo group. For quality evaluation, oped with the aim to reduce the occurrence of adverse we used Cochrane’s risk of bias tool [13]. We evaluated efects associated with fbrate-type drugs and as a drug low risk of bias, moderate risk of bias, and high risk of with fewer restrictions for use in patients with renal dis- bias related to six domains (random sequence generation, ease or for concomitant statin use [10]. Pemafbrate pos- allocation concealment, blinding of personnel and partic- sesses lipid-improving efects similar to those of existing ipants, blinding of outcome assessors, incomplete data, fbrate-type drugs; moreover, its side efects are equiva- and selective reporting). lent to those of the placebo [11, 12]. We considered that highly relevant results related to the efcacy and safety Statistical analysis of pemafbrate administration could be obtained via an As an indicator of the therapeutic efect, the diference integrated analysis of all these previous studies. Tere- between groups was assessed regarding the amount of fore, using a meta-analysis of randomized controlled tri- change of lipid and sugar metabolism-related markers als (RCTs), this study aimed to investigate the efcacy before and after treatment. Since studies used diferent and safety of pemafbrate administration in patients with units, we decided to analyze using standardized mean dyslipidemia. diference (SMD) and 95% CIs. As a safety evaluation index, analysis was performed using odds ratio (OR) Methods and 95% CIs related to total AEs, the increase in hepa- Study selection tobiliary enzyme activities [aspartate aminotransferase Te literature search was performed on November 1, (AST), alanine aminotransferase (ALT), and γ-glutamyl 2018, using MEDLINE (from 1960), Cochrane Controlled transpeptidase (γGTP) activities above the normal upper Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 3 of 14

limit], kidney disorder (creatinine 1.5 mg/dL or more), satisfed the eligibility criteria of this study and were and creatine kinase (CK) increase (above the normal incorporated into the meta-analysis (Fig. 1) [11, 12, 17– upper limit). In cases where only the standard error or 21]. Te characteristics of these seven RCTs are sum- P-values were mentioned, the standard deviation was marized in Table 1. Te dose of pemafbrate in these calculated with reference to Altman and Bland [14]. In studies was 0.025, 0.05, 0.1, 0.2, or 0.4 mg/day. Te com- the absence of the standard deviation, it was calculated parison group was placebo and fenofbrate (100, 106.6, or from 95% CIs, t-values, or P-values [15]. A random efect 200 mg/day). Mean patient age was 52 years, the mean model was used for the analysis, and I­2 was used for the ratio of female patients was 16.9%, and the mean exami- evaluation of the statistical heterogeneity; it was assumed nation period was 16 weeks. 2 that heterogeneity was observed in case of I­ ≥ 50% [16]. In the RCTs included in this study, the proportion of If the number of RCTs incorporated in the analysis was appropriate assessment by each domain was 85.7% (6/7) 10 or more, we would create a funnel plot by evaluation for random sequence generation, 85.7% (6/7) for alloca- of the systematic review [15]. Regarding the comparative tion concealment, 85.7% (6/7) for blinding of participants study of the efects of pemafbrate and placebo on TG, we and personnel, 85.7% (6/7) for blinding of outcome asses- decided to perform subgroup analysis including: (1) base- sors, 85.7% (6/7) for incomplete data and 100% (7/7) for line TG ≥ 300 mg/dL and TG < 300 mg/dL type, (2) pres- selective reporting. Generally speaking, the quality of ence/absence of diabetes type and (3) presence/absence the RCTs included was high. As the number of RCTs of concomitant statin use type. For analysis, RevMan ver- included was less than 10, a funnel plot was not created sion 5.3 (Cochrane Collaboration, http://tech.cochr​ane. (see Additional fle 1: Table S1). org/revma​n/downl​oad, December/2018) was used. Efcacy Results In the study on the efect on TG level, the number of Description of studies included and assessment pooled subjects was 831 in the pemafbrate group and of potential bias 653 in the placebo group. Te statistical heterogeneity 2 In total, 190 papers were extracted from the litera- was ­I = 78% (P < 0.001), indicating signifcant hetero- ture, from which seven RCTs (involving 1623 patients) geneity. In the pemafbrate group, TG level decreased

Fig. 1 Study fow diagram Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 4 of 14 TG (mmol/L) TG 3.4 4.3 346 (mg/dL) 3.2 2.7 3.4 2.9 4 Study Study duration (weeks) 12 24 12 24 24 12 Statin (%) Statin 0 100 0 40.4 0 0 0 Hypertension (%) 25.7 58.5 34.9 37 24.7 18.2 31.3 Diabetes (%) Diabetes 11.4 29.3 16.3 100 6.8 0 12.5 ) 2 BMI (kg/m 26.8 26.2 26.9 26 26.1 25.8 25 % women 2.9 20 12 34 21 0 29 Age (years) Age 48 55 49 61 53 46 52 27 32 No. of patients No. 224 423 526 166 225 Region Japan Japan Japan Japan Japan Japan Japan Year 2016 2017 2018 2018 2018 2018 2018 triglycerides Characteristics of pemafbrate interventions included in the present meta-analysis compared with placebo or fenofbrate interventions or fenofbrate with placebo compared meta-analysis included in the present interventions of pemafbrate Characteristics References al. [ 11 ] Ishibashi et al. al. [ 17 ] Arai et al. al. [ 12 ] Arai et al. Araki [ 18 ] et al. al. [ 19 ] Ishibashi et al. 20 ] [ Matsuba et al. 21 ] [ et al. Yamashita 1 Table as mean values shown otherwise, are Unless indicated data TG BMI body mass index, No. 1 2 3 4 5 6 7 Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 5 of 14

signifcantly compared with the placebo group (SMD, In the study on the infuence on HDL-C level, the num- − 1.38; 95% CI, − 1.63 to − 1.12; P < 0.001; Fig. 2). ber of pooled subjects was 722 in the pemafbrate group Regardless of the dose of pemafbrate, TG in the pemaf- and 539 in the placebo group. Te statistical heterogene- 2 ibrate group decreased signifcantly compared to the ity was I­ = 0% (P = 0.45), meaning that no heterogene- placebo group. On the other hand, there was no sig- ity was observed. In the pemafbrate group, HDL-C level nifcant diference in TG level between the pemafbrate increased signifcantly compared with the placebo group and fenofbrate groups (SMD, − 0.16; 95% CI, − 0.36 to (SMD, 0.77; 95% CI, 0.66–0.89; P < 0.001) (see Addi- 0.03; P = 0.10; Fig. 3). tional fle 1: Fig. S1). Regardless of the dose of pemaf- brate, HDL-C level in the pemafbrate group decreased

Fig. 2 Forest plot presenting meta-analysis based on SMDs for the efect pemafbrate versus placebo on TGs. SMDs in the individual studies are presented as squares with 95% confdence intervals (CIs) presented as extending lines. The pooled SMD with its 95% CI is depicted as a diamond. SMDs, standardized mean diferences; TG, triglycerides Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 6 of 14

Fig. 3 Forest plot presenting meta-analysis based on SMDs for efect of pemafbrate versus fenofbrate on TGs. SMDs in the individual studies are presented as squares with 95% confdence intervals (CIs) presented as extending lines. The pooled SMD with its 95% CI is depicted as a diamond. SMDs, standardized mean diferences; TG, triglycerides signifcantly compared to the placebo group. On the infuence on non-HDL-C level, the number of pooled other hand, there was no signifcant diference in HDL-C subjects was 722 in the pemafbrate group and 539 in the 2 level between the pemafbrate and fenofbrate groups. placebo group. Te statistical heterogeneity was ­I = 0% In the study on the efect on LDL-C level, the number (P = 0.58) and no heterogeneity was observed. Te non- of pooled subjects was 711 in the pemafbrate group and HDL-C level was signifcantly lower in the pemafbrate 532 in the placebo group. Te statistical heterogeneity group than in the placebo group (SMD, 0.39; 95% CI, 2 − was ­I = 28% (P = 0.17), meaning that no heterogene- − 0.51 to − 0.28; P < 0.001) (see Additional fle 1: Fig. S3). ity was observed. In the pemafbrate group, LDL-C level On the other hand, no signifcant diference in non-HDL- increased signifcantly, compared with the placebo group C level was observed between the pemafbrate and fenof- (SMD, 0.19; 95% CI, 0.06–0.33; P = 0.006) (see Additional ibrate groups. fle 1: Fig. S2). Furthermore, in the pemafbrate group, the In the study related to the infuence on homeostasis LDL-C level was signifcantly higher than in the fenof- model assessment for insulin resistance (HOMA-IR), brate group, an efect which was particularly marked at the number of pooled subjects was 585 in the pemaf- the 0.4 mg dose pemafbrate group. In the study on the brate group and 507 in the placebo group. Te statistical Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 7 of 14

2 (see Additional fle 1: Fig. S11), nor between the pemaf- heterogeneity was ­I = 0% (P = 0.94), indicating no het- erogeneity. In the pemafbrate group, the HOMA-IR brate and fenofbrate groups (OR, 0.55; 95% CI, 0.21– was signifcantly lower than in the placebo group (SMD, 1.45; P = 0.23) (see Additional fle 1: Fig. S12). − 0.27; 95% CI, − 0.39 to − 0.14; P < 0.001) (see Addi- tional fle 1: Fig. S4). On the other hand, there was no sig- Discussion nifcant diference in HOMA-IR between the pemafbrate Here, using a meta-analysis of RCTs, we investigated and fenofbrate groups. In the study on the infuence on the efcacy and safety of pemafbrate administration in HbA1c level, the number of pooled subjects was 290 in patients with dyslipidemia. With respect to efcacy, the the pemafbrate group and 238 in the placebo group. Te 2 TG, HDL-C, and non-HDL-C levels as well as HOMA- statistical heterogeneity was ­I = 0% (P = 0.94), indicating IR improved in the pemafbrate group compared with the no heterogeneity. Tere was no signifcant diference in placebo group. In contrast, the LDL-C level was higher HbA1c level between the pemafbrate group and the pla- in the pemafbrate group than in the placebo and fenof- cebo group (SMD, 0.03; 95% CI, − 0.15 to 0.20; P = 0.76) brate groups. With respect to safety, the total AEs were (see Additional fle 1: Fig. S5), while no signifcant difer- not signifcantly diferent between the pemafbrate and ence was observed in HbA1c level between the pemaf- placebo groups; however, an improvement in the hepato- brate group and the fenofbrate group. biliary enzyme activity was observed in the pemafbrate According to the results of subgroup analysis on the group compared with the placebo group. Furthermore, efect on TG level in the pemafbrate and placebo groups, compared with the fenofbrate group, the pemafbrate the TG level decreased signifcantly in the pemafbrate group showed an elevated hepatobiliary enzyme activity, group compared with the placebo group, regardless of the total AEs were signifcantly fewer, and the increase in the baseline TG value, the presence or absence of diabe- Cre level was signifcantly lower. tes and the presence or absence of the concomitant use of When PPARα is activated in the liver, the expression statin (see Additional fle 1: Figs. S6–S8). of apolipoprotein C-III, which inhibits the activity of the TG-hydrolyzing lipoprotein lipase (LPL) enzyme, Safety decreases, thereby causing increased LPL activity that Investigation of total AEs indicated no signifcant dif- leads to decreased blood TG levels [22]. Furthermore, the ference in total AEs between the pemafbrate group and activation of PPARα induces the expression of apolipo- the placebo group (OR, 0.92; 95% CI, 0.74–1.14; P = 0.45; protein A-I and A-II, which are constituents of HDL-C, Fig. 4). On the other hand, total AEs in the pemafbrate such that the HDL-C level increases [23]. In the present group was signifcantly lower than in the fenofbrate study, compared with the placebo group, although the group (OR, 0.60; 95% CI, 0.49–0.73; P < 0.001; Fig. 5). Te pemafbrate group showed a decrease in the TG level and frequency of hepatobiliary enzyme activity increase was an increase in the HDL-C level, there was no signifcant signifcantly lower in the pemafbrate group as compared diference with the levels in the fenofbrate group. How- with either the placebo group (OR, 0.33; 95% CI, 0.21– ever, with respect to safety, the frequency of hepatobiliary 0.52; P < 0.001; Fig. 6) or the fenofbrate group (OR, 0.14; enzyme activity increase was lower in the pemafbrate 95% CI, 0.10–0.20; P < 0.001; Fig. 7). In particular, ALT group than in the placebo group and fenofbrate groups; activity decreased signifcantly following administration indeed, an improvement in hepatobiliary enzyme activ- of pemafbrate as compared with the placebo group while ity was observed in the pemafbrate group. Te activation γGTP activity decreased signifcantly as compared with of PPARα by pemafbrate reportedly accelerates fatty acid both the placebo group and the fenofbrate group. oxidation, which in turn promotes energy combustion in Investigation of kidney dysfunction indicated no difer- the liver and reduces fat accumulation [22, 23]. ence in the frequency of kidney dysfunction when com- In the present study, ALT and γGTP activities were paring the pemafbrate group and the placebo group (OR, lower in the pemafbrate group than in the placebo group 1.67; 95% CI, 0.63–4.42; P = 0.30) (see Additional fle 1: and γGTP activity decreased compared with the fenof- Fig. S9), or when comparing with the fenofbrate group brate group; this ALT-improving efect was particularly (OR, 0.51; 95% CI, 0.20–1.32; P = 0.19) (see Additional marked in the groups administered the highest doses (0.2 fle 1: Fig. S10). However, although Cre level increased and 0.4 mg/day) of pemafbrate, suggesting the likelihood in the pemafbrate group compared with the placebo of a dose-dependent efect. In animal experiments, fatty group, the increase in Cre level in the pemafbrate group liver improvement, liver fbrosis suppression, and nonal- was signifcantly lower than in the fenofbrate group. No coholic steatohepatitis improvement have been reported signifcant diference was found in the frequency of CK following the administration of fbrate-type drugs [24, activity increase between the pemafbrate group and the 25]. Considering it is difcult to continue the adminis- placebo group (OR, 0.50; 95% CI, 0.24–1.05; P = 0.07) tration of fbrate-type drugs because of hepatic disorder Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 8 of 14

Fig. 4 Forest plot presenting meta-analysis on OR for efect of pemafbrate versus placebo on total AEs. OR in the individual studies are presented as squares with 95% confdence intervals (CIs) presented as extending lines. The pooled OR with its 95% CI is depicted as a diamond. OR, odds ratio; AEs, adverse events Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 9 of 14

Fig. 5 Forest plot presenting meta-analysis on OR for efect of pemafbrate versus fenofbrate on total AEs. OR in the individual studies are presented as squares with 95% confdence intervals (CIs) presented as extending lines. The pooled OR with its 95% CI is depicted as a diamond. OR, odds ratio; AEs, adverse events

development [8], pemafbrate may likely be useful from Although there were no diferences in the HbA1c the viewpoint of its benefcial efects on the liver and its levels in the pemafbrate group compared with the pla- tolerability by the patient. cebo or fenofbrate groups, HOMA-IR decreased in the Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 10 of 14

Fig. 6 Forest plot presenting meta-analysis on OR for efect of pemafbrate versus placebo on hepatobiliary enzyme. OR in the individual studies are presented as squares with 95% confdence intervals (CIs) presented as extending lines. The pooled OR with its 95% CI is depicted as a diamond. OR, odds ratio Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 11 of 14

Fig. 7 Forest plot presenting meta-analysis on OR for efect of pemafbrate versus fenofbrate on hepatobiliary enzyme. OR in the individual studies are presented as squares with 95% confdence intervals (CIs) presented as extending lines. The pooled OR with its 95% CI is depicted as a diamond. OR, odds ratio

pemafbrate group compared with the placebo group. in the liver, promotion of LPL production, and promo- Activation of PPARα may lead to decreases in the tion of apolipoprotein A-I and A-II production [22, 23]. hepatic fat mass owing to the promotion of β-oxidation Increases in the hepatic fat mass, decreases in the TG Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 12 of 14

levels, and increases in the HDL-C levels are associated by patients. PPARα assumedly has multifaceted efects, with decreases in HOMA-IR [26]. Although the efects such as anti-infammatory and antioxidant activities, in of PPARα agonists (other than pemafbrate) on HOMA- addition to its efect on lipids and has been proposed IR have been reported in humans, the consistency of the to suppress arteriosclerotic disease onset [34, 35]. A results is low [27–29]. In contrast, the results of the pre- recent study has suggested the involvement of angi- sent study corroborate with those reported by previous opoietin-like protein 8 (ANGPTL8) in the association studies on improved insulin resistance following pemaf- between dyslipidemia and arteriosclerosis [36], indicat- brate administration [20, 30]. How insulin resistance ing that hypertriglyceridemia is partially involved in improvement in response to pemafbrate administra- the association between ANGPTL8 and arteriosclero- tion infuences sugar metabolism, lipid metabolism, and sis [36]. A study has reported that ANGPTL8 may be cardiovascular disease onset needs to be investigated in involved in HDL-C dysfunction [37], and another study future studies. has indicated the close association of ANGPTL8 with In the present study, the LDL-C levels were higher in dyslipidemia regardless of glucose intolerance or dia- the pemafbrate group than in the placebo and fenof- betes mellitus [38]. Taken together, these reports sug- brate groups. As mentioned earlier, the activation of gest that lipid management, including TG and HDL-C, PPARα increases the LPL levels and decreases the TG is extremely important in the prevention of cardio- levels [22]. As with other fbrate-type drugs, the LDL-C vascular diseases. A large-scale clinical trial [entitled levels assumedly increase as a result of very low-density “Pemafbrate to Reduce Cardiovascular Outcomes lipoproteins containing large amounts of apolipopro- by Reducing Triglycerides in Patients with Diabetes tein B being converted to LDL-C owing to the catabolic (PROMINENT)”] assessing the efect of pemafbrate action of pemafbrate [31, 32]. However, a previous administration on cardiovascular outcomes is under- study reported that LDL-C closely related to arterioscle- way [39], and the results of this trial are eagerly awaited. rotic disease onset was markedly reduced by pemaf- Te present study has several limitations. First, the brate administration, whereas the increased LDL-C type RCTs included in this study involved only data from ranged from medium to large LDL-C [11]. Terefore, clinical trials. In the future, research data, including clini- although the increase in the LDL-C levels following cal research data, may have to be re-examined. Second, pemafbrate administration is expected to be safe, further given the small number of RCTs included in the cur- future investigations are required to elucidate the clini- rent meta-analysis, it is possible that some relevant data cal signifcance of this efect. In contrast, with respect to may not have been searched this time, thereby infuenc- the efects of pemafbrate administration on the Cre level, ing the results and causing biases. Tird, this study only which was lower in the pemafbrate group than in the in included Japanese subjects; therefore, the results of this the fenofbrate group, there was a slight increase com- study cannot be generalized to other populations. Finally, pared with the placebo group. Many fbrate-type drugs, the observation period of the RCTs included in this study including fenofbrate, are renal excretory and may cause was comparatively short and the results cannot be related renal disorders [9]. In contrast, because pemafbrate is to long-term efcacy and safety. Terefore, the results of biliary excretory, it can hardly cause kidney damage [33]; the present study should be interpreted with caution. this diference in pharmacokinetics has been inferred to be the cause of the results of the present study. Com- pared with placebo, although pemafbrate increases the Conclusions Cre level to a very small extent, it seems necessary to In summary, using a meta-analysis of RCTs, we inves- carefully consider the potential risks as well as benefts of tigated the efcacy and safety of pemafbrate adminis- pemafbrate. tration in patients with dyslipidemia. Te lipid profle Pemafbrate has been developed with an aim of signifcantly improved in the pemafbrate group than in reducing TG, which is reported as a risk factor for the placebo group. Apart from the fact that the pemaf- development of cardiovascular diseases in patients brate group showed a lipid-improving efect equiva- with dyslipidemia or insulin resistance, or increasing lent to that in the fenofbrate group, the total AEs was HDL-C levels [10]. Te present study demonstrated signifcantly fewer in the pemafbrate group than in the that the lipid-improving efect of pemafbrate is com- fenofbrate group and the hepatobiliary enzyme activity parable with that of fenofbrate. Furthermore, pemaf- was actually improved. However, the clinical data of daily brate showed fewer increases in total AEs and a smaller medical practice and the long-term efcacy and safety increase in hepatobiliary system enzyme activities of pemafbrate administration need to be investigated in than that showed by fenofbrate; these efects assum- future studies. edly contribute to the high tolerability of pemafbrate Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 13 of 14

Additional fle cholesterol to residual risk of coronary heart disease after establish- ment of low-density lipoprotein cholesterol control. Am J Cardiol. 2010;106:757–63. Additional fle 1. Additional fgures and tables. 5. Chapman MJ, Redfern JS, McGovern ME, Giral P. Niacin and fbrates in atherogenic dyslipidemia: pharmacotherapy to reduce cardiovascular risk. Pharmacol Ther. 2010;126:314–45. Abbreviations 6. Jun M, Foote C, Lv J, Neal B, Patel A, Nicholls SJ, et al. Efects of fbrates CI: confdence interval; RCTs: randomized controlled trials; SMD: standardized on cardiovascular outcomes: a systematic review and meta-analysis. mean diference; OR: odds ratio; TG: triglycerides; HDL-C: high-density lipo- Lancet. 2010;375:1875–84. protein cholesterol; LDL-C: low-density lipoprotein cholesterol; Non-HDL-C: 7. Wang D, Liu B, Tao W, Hao Z, Liu M. Fibrates for secondary prevention non-high-density lipoprotein cholesterol; HbA1c: hemoglobin A1c; HOMA-IR: of cardiovascular disease and stroke. Cochrane Database Syst Rev. homeostasis model assessment for insulin resistance; ALT: alanine aminotrans- 2015;10:CD009580. ferase; ALT: alanine aminotransferase; γGTP: γ-glutamyl transpeptidase; AEs: 8. Catapano AL, Graham I, De Backer G, Wiklund O, Chapman MJ, Drexel adverse events; PPARα: peroxisome proliferator-activated receptor α; SPPARMα: H, et al. 2016 ESC/EAS Guidelines for the Management of Dyslipidae- selective peroxisome proliferator-activated receptor α; CK: creatine kinase; LPL: mias: the Task Force for the Management of Dyslipidaemias of the lipoprotein lipase. European Society of Cardiology (ESC) and European Atherosclerosis Society (EAS) Developed with the special contribution of the European Authors’ contributions Assocciation for Cardiovascular Prevention & Rehabilitation (EACPR). SI designed the study and drafted the manuscript. KM interpreted the result Atherosclerosis. 2016;253:281–344. data and reviewed them from a medical perspective. RK assisted in drafting 9. Davidson MH, Armani A, McKenney JM, Jacobson TA. Safety considera- the manuscript. All authors read and approved the fnal manuscript. tions with fbrate therapy. Am J Cardiol. 2007;99:S3–18. 10. Fruchart JC. Pemafbrate (K-877), a novel selective peroxisome proliferator-activated receptor alpha modulator for management of Acknowledgements atherogenic dyslipidaemia. Cardiovasc Diabetol. 2017;16:124. The authors would like to thank the staf members of the Department of 11. Ishibashi S, Yamashita S, Arai H, Araki E, Yokote K, Suganami H, et al. Metabolic Diseases at the Ise Red Cross Hospital for their cooperation in this Efects of K-877, a novel selective PPARalpha modulator (SPPARMal- study. pha), in dyslipidaemic patients: a randomized, double blind, active- and placebo-controlled, phase 2 trial. Atherosclerosis. 2016;249:36–43. Competing interests 12. Arai H, Yamashita S, Yokote K, Araki E, Suganami H, Ishibashi S, et al. The authors declare that they have no competing interests. Efcacy and safety of pemafbrate versus fenofbrate in patients with high triglyceride and low hdl cholesterol levels: a multicenter, placebo- Availability of data and materials controlled, double-blind, randomized trial. J Atheroscler Thromb. The data that support the fndings of this study are available from the cor- 2018;25:521–38. responding author upon reasonable request. 13. Simmonds MC, Higgins JP, Stewart LA, Tierney JF, Clarke MJ, Thompson SG. Meta-analysis of individual patient data from randomized trials: a Consent for publication review of methods used in practice. Clin Trials. 2005;2:209–17. Not applicable. 14. Altman DG, Bland JM. Detecting skewness from summary information. BMJ. 1996;9(313):1200. Ethics approval and consent to participate 15. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Not applicable. Interventions Version 5.1.0. The Cochrane Collaboration. 2011. http:// www.cochr​ane-handb​ook.org. Accessed 1 Nov 2018. Funding 16. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsist- Not applicable. ency in meta-analyses. BMJ. 2003;327:557–60. 17. Arai H, Yamashita S, Yokote K, Araki E, Suganami H, Ishibashi S, et al. Efcacy and safety of K-877, a novel selective peroxisome proliferator- Publisher’s Note activated receptor alpha modulator (SPPARMalpha), in combination Springer Nature remains neutral with regard to jurisdictional claims in pub- with statin treatment: two randomised, double-blind, placebo- lished maps and institutional afliations. controlled clinical trials in patients with dyslipidaemia. Atherosclerosis. 2017;261:144–52. Received: 30 January 2019 Accepted: 13 March 2019 18. Araki E, Yamashita S, Arai H, Yokote K, Satoh J, Inoguchi T, et al. Efects of pemafbrate, a novel selective pparalpha modulator, on lipid and glucose metabolism in patients with type 2 diabetes and hypertriglyc- eridemia: a randomized, double-blind, placebo-controlled, phase 3 trial. Diabetes Care. 2018;41:538–46. References 19. Ishibashi S, Arai H, Yokote K, Araki E, Suganami H, Yamashita S, et al. 1. Stone NJ, Robinson JG, Lichtenstein AH, Bairey Merz CN, Blum CB, Eckel Efcacy and safety of pemafbrate (K-877), a selective peroxisome RH, et al. 2013 ACC/AHA guideline on the treatment of blood choles- proliferator-activated receptor alpha modulator, in patients with terol to reduce atherosclerotic cardiovascular risk in adults: a report of dyslipidemia: results from a 24-week, randomized, double blind, active- the American College of Cardiology/American Heart Association Task controlled, phase 3 trial. J Clin Lipidol. 2018;12:173–84. Force on Practice Guidelines. J Am Coll Cardiol. 2014;63:2889–934. 20. Matsuba I, Matsuba R, Ishibashi S, Yamashita S, Arai H, Yokote K, et al. 2. Teramoto T, Sasaki J, Ishibashi S, Birou S, Daida H, Dohi S, et al. Execu- Efects of a novel selective peroxisome proliferator-activated receptor- tive summary of the Japan Atherosclerosis Society (JAS) guidelines for alpha modulator, pemafbrate, on hepatic and peripheral glucose the diagnosis and prevention of atherosclerotic cardiovascular diseases uptake in patients with hypertriglyceridemia and insulin resistance. J in Japan-2012 version. J Atheroscler Thromb. 2013;20:517–23. Diabetes Investig. 2018;9:1323–32. 3. Cholesterol Treatment Trialists’ (CTT) Collaboration, Baigent C, 21. Yamashita S, Arai H, Yokote K, Araki E, Suganami H, Ishibashi S, et al. Blackwell L, Emberson J, Holland LE, Reith C, et al. Efcacy and safety Efects of pemafbrate (K-877) on cholesterol efux capacity and of more intensive lowering of LDL cholesterol: a meta-analysis of postprandial hyperlipidemia in patients with atherogenic dyslipidemia. data from 170,000 participants in 26 randomised trials. Lancet. J Clin Lipidol. 2018;12:1267–79. 2010;376:1670–81. 22. Staels B, Dallongeville J, Auwerx J, Schoonjans K, Leitersdorf E, Fruchart 4. Carey VJ, Bishop L, Laranjo N, Harshfeld BJ, Kwiat C, Sacks FM. Contri- JC. Mechanism of action of fbrates on lipid and lipoprotein metabolism. bution of high plasma triglycerides and low high-density lipoprotein Circulation. 1998;98:2088–93. Ida et al. Cardiovasc Diabetol (2019) 18:38 Page 14 of 14

23. Vu-Dac N, Schoonjans K, Laine B, Fruchart JC, Auwerx J, Staels B. Negative 31. Sigurdsson G, Nicoll A, Lewis B. Conversion of very low density lipopro- regulation of the human apolipoprotein A-I promoter by fbrates can be tein to low density lipoprotein. A metabolic study of apolipoprotein B attenuated by the interaction of the peroxisome proliferator-activated kinetics in human subjects. J Clin Invest. 1975;56:1481–90. receptor with its response element. J Biol Chem. 1994;269:31012–8. 32. Davidson MH, Rosenson RS, Maki KC, Nicholls SJ, Ballantyne CM, Maz- 24. Honda Y, Kessoku T, Ogawa Y, Tomeno W, Imajo K, Fujita K, et al. Pemaf- zone T, et al. Efects of fenofbric acid on carotid intima-media thickness brate, a novel selective peroxisome proliferator-activated receptor alpha in patients with mixed dyslipidemia on atorvastatin therapy: rand- modulator, improves the pathogenesis in a rodent model of nonalcoholic omized, placebo-controlled study (FIRST). Arterioscler Thromb Vasc Biol. steatohepatitis. Sci Rep. 2017;7:42477. 2014;34:1298–306. 25. Staels B, Rubenstrunk A, Noel B, Rigou G, Delataille P, Millatt LJ, et al. 33. Blair HA. Pemafbrate: frst global approval. Drugs. 2017;77:1805–10. Hepatoprotective efects of the dual peroxisome proliferator-activated 34. Sahebkar A, Hernandez-Aguilera A, Abello D, Sancho E, Camps J, Joven J. receptor alpha/delta agonist, GFT505, in rodent models of nonalco- Systematic review and meta-analysis deciphering the impact of fbrates holic fatty liver disease/nonalcoholic steatohepatitis. Hepatology. on paraoxonase-1 status. Metabolism. 2016;65:609–22. 2013;58:1941–52. 35. Ghani RA, Bin Yaakob I, Wahab NA, Zainudin S, Mustafa N, Sukor N, et al. 26. Robins SJ, Lyass A, Zachariah JP, Massaro JM, Vasan RS. Insulin resistance The infuence of fenofbrate on lipid profle, endothelial dysfunction, and and the relationship of a dyslipidemia to coronary heart disease: the infammatory markers in type 2 diabetes mellitus patients with typical Framingham Heart Study. Arterioscler Thromb Vasc Biol. 2011;31:1208–14. and mixed dyslipidemia. J Clin Lipidol. 2013;7:446–53. 27. Bajaj M, Suraamornkul S, Hardies LJ, Glass L, Musi N, DeFronzo RA. Efects 36. Zheng T, Ge B, Liu H, Chen B, Qin L, Xiao L, et al. Triglyceride-mediated of peroxisome proliferator-activated receptor (PPAR)-alpha and PPAR- infuence of serum angiopoietin-like protein 8 on subclinical atheroscle- gamma agonists on glucose and lipid metabolism in patients with type 2 rosis in type 2 diabetic patients: results from the GDMD study in China. diabetes mellitus. Diabetologia. 2007;50:1723–31. Cardiovasc Diabetol. 2018;17:84. 28. Cariou B, Hanf R, Lambert-Porcheron S, Zair Y, Sauvinet V, Noel B, et al. 37. Luo M, Zhang Z, Peng Y, Wang S, Peng D. The negative efect of ANGPTL8 Dual peroxisome proliferator-activated receptor alpha/delta agonist on HDL-mediated cholesterol efux capacity. Cardiovasc Diabetol. GFT505 improves hepatic and peripheral insulin sensitivity in abdomi- 2018;17:142. nally obese subjects. Diabetes Care. 2013;36:2923–30. 38. Gomez-Ambrosi J, Pascual-Corrales E, Catalan V, Rodriguez A, Ramirez B, 29. Black RN, Ennis CN, Young IS, Hunter SJ, Atkinson AB, Bell PM. The peroxi- Romero S, et al. Altered concentrations in dyslipidemia evidence a role for some proliferator-activated receptor alpha agonist fenofbrate has no ANGPTL8/Betatrophin in lipid metabolism in humans. J Clin Endocrinol efect on insulin sensitivity compared to atorvastatin in type 2 diabetes Metab. 2016;101:3803–11. mellitus; a randomised, double-blind controlled trial. J Diabetes Compl. 39. Pradhan AD, Paynter NP, Everett BM, Glynn RJ, Amarenco P, Elam M, et al. 2014;28:323–7. Rationale and design of the pemafbrate to reduce cardiovascular out- 30. Araki E, Ishibashi S, Yamashita S, Arai H, Yokote K, Suganami H, et al. A comes by reducing triglycerides in patients with diabetes (PROMINENT) highly potent and specifc PPAR alpha agonist, K-877, improves lipid study. Am Heart J. 2018;206:80–93. profles and insulin sensitivity in dyslipidaemia subjects; an integrated analysis of 3 phase 2/3 trials (Abstract). Diabetologia. 2014;57:S272.

Ready to submit your research ? Choose BMC and benefit from:

• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions