Efficacy and Safety of Bococizumab (RN316/PF-04950615), a Monoclonal Antibody Against Proprotein Convertase Subtilisin/Kexin Type 9, in Hypercholesterolemic Japanese Subjects Receiving a Stable Dose of Atorvastatin or Treatment-Naive　― Results From a Randomized, Placebo-Controlled, Dose-Ranging Study ―

Koutaro Yokote; Shigeto Kanada; Osamu Matsuoka; Hisakuni Sekino; Keiji Imai; Junichi Tabira; Nobushige Matsuoka; Sandip Chaudhuri; Tamio Teramoto

doi:10.1253/circj.CJ-16-1310

Abstract

Background: A Phase 2, dose-ranging study of bococizumab, a monoclonal anti-proprotein convertase subtilisin/kexin type 9 antibody, was conducted in Japanese subjects to assess its efficacy, safety, and tolerability in this population.

Methods and Results: Two different hypercholesterolemic study populations were enrolled concurrently: Japanese subjects with uncontrolled low-density lipoprotein cholesterol (LDL-C) despite atorvastatin treatment (LDL-C ≥100 mg/dL; n=121), and Japanese subjects naive to lipid-lowering agents and with LDL-C ≥130 mg/dL (n=97). Subjects within each study population were randomized to bococizumab 50, 100, or 150 mg, or placebo, q14D for 16 weeks; an open-label ezetimibe 10 mg daily arm was also included for the atorvastatin-treated population. Significant, dose-dependent reductions in fasting LDL-C levels were observed in all bococizumab arms of both study populations at Weeks 12 and 16 (adjusted mean percent changes from baseline: 54.1–76.7% for atorvastatin-treated subjects and 47.7–66.8% for treatment-naive subjects; P<0.001 vs. placebo for all). Bococizumab also caused dose-dependent changes in other lipid parameters in both study populations at Weeks 12 and 16. No serious adverse events (AEs) related to bococizumab treatment occurred and all treatment-emergent AEs were mild or moderate in severity. No dose-dependent relationship between bococizumab treatment and development of anti-drug antibodies was observed.

Conclusions: Bococizumab was well tolerated and significantly reduced fasting LDL-C in atorvastatin-treated and treatment-naive hypercholesterolemic Japanese subjects. (Clinicaltrials.gov identifier: NCT02055976.)

Hypercholesterolemia caused by high levels of low-density lipoprotein cholesterol (LDL-C) has been demonstrated by numerous epidemiologic studies to be a risk factor for atherosclerotic cardiovascular (CV) disease, including myocardial infarction and ischemic stroke.¹^,² Furthermore, numerous prospective clinical outcome trials and meta-analyses have established that lowering LDL-C decreases CV morbidity and mortality.³

Editorial p ????

Proprotein convertase subtilisin/kexin type 9 (PCSK9) is involved in regulation of serum LDL-C levels via its binding to, and downregulation of, the LDL receptor (LDLR) expressed on hepatocytes. This reduction in cell-surface LDLR results in reduced hepatocellular uptake of LDL-C and, consequently, higher LDL-C levels within the circulation. Bococizumab (RN316/PF-04950615) is a humanized monoclonal anti-PCSK9 antibody with high affinity for the evolutionarily conserved LDLR-binding domain of PCSK9.⁴ Bococizumab has been investigated worldwide as an add-on therapy to the standard of care in patients with hypercholesterolemia or heterozygous familial hypercholesterolemia. Phase 1 and Phase 2a studies have shown that bococizumab is well tolerated and provides substantial reductions in fasting LDL-C levels when administered either as single or multiple doses, both alone and in combination with other lipid-lowering agents.⁵^,⁶ Based on the results of a global Phase 2b study⁷ and a modeling simulation of the available pharmacokinetic (PK) and pharmacodynamic (PD) data,⁸ a twice-monthly (q14D) regimen is considered the preferred dosing frequency for bococizumab.⁷

The current Phase 2 study conducted in Japan (NCT02055976) evaluated the efficacy, safety, and tolerability of bococizumab when administered subcutaneously q14D.

Methods

Subjects

Men and women aged ≥20 years and able to comply with all study procedures were enrolled and included if they, based on the Japan Atherosclerosis Society guidelines for prevention of atherosclerotic CV diseases,² displayed fasting LDL-C ≥100 mg/dL and fasting triglycerides (TG) ≤400 mg/dL despite taking a stable dose of atorvastatin (‘atorvastatin-treated subjects’), or if they had not received any lipid-lowering agents and displayed fasting LDL-C and TG levels of ≥130 mg/dL and ≤400 mg/dL, respectively (‘treatment-naive subjects’). Greater detail on inclusion and exclusion criteria can be found in Supplementary File 1.

Study Design

This was a double-blind, parallel-group, placebo-controlled, randomized, dose-ranging trial conducted in Japan and involving 2 different study populations with a treatment duration of 16 weeks (4 months). A total of 9 parallel treatment arms with 24 subjects per arm was planned (Figure 1). The atorvastatin plus ezetimibe 10 mg daily arm was open-label, whereas the 8 other arms were blinded. Institutional review board(s) and/or independent ethics committee(s) at each of the 9 participating centers approved the protocol, and all subjects provided written informed consent before initiation of study-specific procedures. The study was conducted in accordance with the principles set forth in the International Ethical Guidelines for Biomedical Research Involving Human Subjects, the Guidelines for Good Clinical Practice, and the Declaration of Helsinki.

Figure 1.

(A,B) Subject disposition in the 2 study populations. AE, adverse event; LDL-C, low-density lipoprotein cholesterol.

A computer-generated randomization schedule was used to assign a study drug in each of the 2 study populations, with subjects in each population stratified into ‘low LDL-C’ (≤130 mg/dL for atorvastatin-treated subjects and ≤150 mg/dL for treatment-naive subjects) or ‘high LDL-C’ (>130 mg/dL for atorvastatin-treated subjects and >150 mg/dL for treatment-naive subjects) groups prior to randomization.

The participating subjects, investigators, site staff, and the sponsor (excluding prespecified, unblinded site staff such as unblinded pharmacists involved in study drug preparation) remained blinded to the assigned study drug and lipid results (except TG) for the duration of the study.

Study Treatment

The doses of bococizumab selected for this study were based on a global Phase 2b dose-ranging study and PK/PD modeling.⁷^,⁸ During the treatment period (Days 1–113), subjects randomized to bococizumab received their assigned dose by subcutaneous administration q14D, whereas those randomized to open-label ezetimibe took a daily 10-mg tablet after food. Subjects received the assigned dose of bococizumab for the first 12 weeks, with dose adjustment on or after Week 12 (Day 85) for any subject who met the dose-adjustment criteria. A subject’s scheduled dose of bococizumab was adjusted if they displayed 2 consecutive LDL-C measurements ≤25 mg/dL at the end of the dosing interval at Weeks 8, 10, and 12, with the dose being adjusted at Week 12 (Day 85) or Week 14 (Day 99) as follows: bococizumab 150 mg→75 mg; bococizumab 100 mg→50 mg; bococizumab 50 mg→placebo. All subjects in the atorvastatin-treated population continued to self-administer their daily dose of atorvastatin as background therapy throughout the study period. Only subjects without changes in their daily dose of atorvastatin for at least 6 weeks prior to screening were included, and each subject’s daily atorvastatin dose could not be adjusted during the trial unless an associated serious adverse event (SAE) occurred. The atorvastatin-treated subjects were also allowed to remain on stable doses of lipid-lowering drugs other than atorvastatin, ezetimibe, and fibrates (e.g., resins, niacin, and omega-3 fatty acids) if these had already been initiated and if the dose had remained unchanged for at least 6 weeks prior to screening. No dose adjustments for these concomitant lipid-lowering medications were allowed during the trial.

Endpoints

The co-primary efficacy endpoints were the percent change from baseline in fasting LDL-C at Week 12 (Day 85) and Week 16 (Day 113). Secondary efficacy endpoints included the absolute value and absolute change from baseline in LDL-C at Weeks 12 and 16, as well as the absolute value, absolute change, and percent change from baseline in other serum lipid parameters (total cholesterol, high-density lipoprotein cholesterol [HDL-C], TG, non-HDL-C, total cholesterol/HDL-C ratio, apolipoprotein B [apoB], apoA-I, apoA-II, lipoprotein(a) [Lp(a)], very low-density lipoprotein cholesterol [VLDL-C], and apoB/apoA-I ratio) and the proportion of subjects achieving specific LDL-C values during the treatment period (<100, <70, <40, <25, and <10 mg/dL). Safety endpoints included the incidence of adverse events (AEs), clinical laboratory abnormalities, and anti-drug antibodies (ADAs).

Laboratory Methods

Serum LDL-C measurements used for dose adjustments and analyzing lipid-lowering effects were calculated by the Friedewald formula, with values obtained by direct measurement used if the calculated value was <25 mg/dL or if TG were ≥400 mg/dL.

Statistical Analysis

The full analysis set was the primary analysis set for the analyses of efficacy data and included all randomized subjects administered ≥1 dose of study treatment. For all efficacy analyses, subjects were analyzed according to their randomized dose regardless of any subsequent dose reductions. The safety analysis set included all subjects administered ≥1 dose of study treatment. All efficacy and safety analyses were conducted in each study population separately. The co-primary endpoints were compared with the corresponding placebo arm using a mixed-effects model for repeated measures (MMRM). The secondary efficacy endpoints were analyzed using the same MMRM analysis. For the statistical test of the co-primary endpoints, multiplicity was adjusted to control family-wise type I error at α=0.025 (one-sided). The model-adjusted mean percent changes from baseline are reported. Data from the open-label ezetimibe arm were summarized descriptively. AEs and SAEs were summarized using the ICH Medical Dictionary for Regulatory Activities (MedDRA) version 17.1. Safety laboratory parameters and incidence of positive ADA results were summarized.

Results

Subject Disposition

Atorvastatin-Treated Subjects A total of 121 subjects were randomized across 5 treatment arms and all received at least 1 dose of study treatment (Figure 1A). Of the subjects receiving bococizumab, dose titrations according to the prespecified algorithm occurred in 1 subject (4%) in the 50-mg arm, 8 subjects (33%) in the 100-mg arm, and 8 subjects (33%) in the 150-mg arm.

Treatment-Naive Subjects A total of 97 subjects were randomized across the 4 treatment arms and all received at least 1 dose of study treatment (Figure 1B). Of the subjects receiving bococizumab, dose titrations according to the prespecified algorithm occurred in 3 subjects (13%) in the 150-mg arm.

Baseline Demographics

All subjects were ethnically Japanese, the age range was 26–80 years, and 54% were men. The baseline characteristics within each study population were similar across the treatment arms (Table 1). Baseline mean LDL-C concentrations ranged from 123.85 to 135.90 mg/dL in the atorvastatin-treated subjects, and from 155.22 to 164.22 mg/dL in the treatment-naive subjects.

Table 1. Baseline Demographics of Atorvastatin-Treated and Treatment-Naive Japanese Study Subjects

Variable^†	Atorvastatin-treated subjects*					Treatment-naive subjects
	Placebo (n=26)	Bococizumab			Ezetimibe 10 mg (n=22)	Placebo (n=23)	Bococizumab
	Placebo (n=26)	50 mg (n=25)	100 mg (n=24)	150 mg (n=24)	Ezetimibe 10 mg (n=22)	Placebo (n=23)	50 mg (n=25)	100 mg (n=25)	150 mg (n=24)
Age (years)	58.2±11.2	55.9±8.9	59.6±12.2	59.6±8.6	58.2±10.7	58.2±11.7	60.1±8.4	57.7±10.6	53.3±8.1
Male sex	13 (50.0)	17 (68.0)	14 (58.3)	12 (50.0)	12 (54.5)	13 (56.5)	12 (48.0)	10 (40.0)	15 (62.5)
Weight (kg)	67.6±13.4	67.7±13.0	66.4±15.3	63.5±11.6	66.6±10.9	63.4±14.0	65.3±12.5	65.9±11.9	68.0±17.0
BMI (kg/m²)	25.6±3.3	24.4±3.4	24.9±4.1	24.2±3.0	24.7±2.6	23.7±3.9	24.9±3.1	25.6±3.8	24.7±5.1
CVD risk factors present^‡
Hypertension	13 (50.0)	8 (32.0)	12 (50.0)	10 (41.7)	8 (36.4)	8 (34.8)	9 (36.0)	8 (32.0)	8 (33.3)
DM	4 (15.4)	3 (12.0)	9 (37.5)	7 (29.2)	4 (18.2)	3 (13.0)	4 (16.0)	3 (12.0)	2 (8.3)
CAD	1 (3.8)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
Stroke	0 (0.0)	1 (4.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
PAD	0 (0.0)	0 (0.0)	1 (4.2)	0 (0.0)	1 (4.5)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
CKD	2 (7.7)	3 (12.0)	3 (12.5)	2 (8.3)	4 (18.2)	1 (4.3)	0 (0.0)	1 (4.0)	1 (4.2)
LDL-C (mg/dL)	135.90±24.70	135.36±23.65	123.85±20.59	129.19±17.77	135.36±24.75	155.22±23.10	164.22±25.84	158.00±20.00	159.90±19.80
Total cholesterol (mg/dL)	222.44±31.32	214.16±30.66	210.29±21.59	211.71±23.07	222.84±25.40	242.04±22.96	245.94±30.43	243.26±21.97	244.19±29.99
HDL-C (mg/dL)	58.54±14.68	54.58±9.05	54.75±11.79	57.00±12.86	56.98±9.61	60.87±10.61	57.22±14.15	60.40±13.23	62.02±21.46
Triglycerides (mg/dL)	128.50 (106.00, 185.50)	107.50 (86.00, 138.00)	141.00 (112.75, 180.75)	114.25 (85.25, 152.25)	156.25 (110.00, 177.00)	125.00 (99.00, 151.00)	111.50 (83.00, 159.00)	109.50 (95.50, 137.00)	96.75 (87.75, 139.00)
Non-HDL-C (mg/dL)	163.90±25.27	159.58±29.03	155.54±20.71	154.71±22.51	165.86±26.66	181.17±23.58	188.72±30.22	182.86±22.84	182.17±21.22
PCSK9 (ng/mL)	269.19±66.84	266.64±61.55	261.23±72.36	285.96±75.79	–	210.22±48.55	228.80±50.00	215.50±43.01	234.42±77.54

*Hypercholesterolemic subjects with fasting LDL-C ≥100 mg/dL and not controlled by stable dose of atorvastatin. ^†Values are mean±standard deviation or n (%), except values for triglycerides, which are presented as median (Q1, Q3). ^‡According to the coding definitions of the ICH Medical Dictionary for Regulatory Activities (MedDRA) version 17.1. BMI, body mass index; CAD, coronary artery disease; CKD, chronic kidney disease; CVD, cardiovascular disease; DM, diabetes mellitus; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; PAD, peripheral arterial disease; PCSK9, proprotein convertase subtilisin/kexin type 9.

Efficacy

Atorvastatin-Treated Subjects Significant dose-dependent reductions in fasting LDL-C levels were observed in all bococizumab treatment arms compared with placebo at both Week 12 and Week 16 (adjusted mean percent reductions of 54.1–76.7%; P<0.001 for all 3 bococizumab arms at both time points; Figure 2A). The adjusted mean absolute reductions in fasting LDL-C levels observed in the bococizumab arms were also dose-dependent and significant compared with placebo at Week 12 and Week 16 (P<0.001 for all 3 arms at both time points; Table S1). The down-titration of dose on or after Week 12 was reflected in the LDL-C-lowering effects observed in the bococizumab 100-mg and 150-mg arms at Week 16 being slightly reduced compared with those observed at Week 12. Dose-dependent reductions in fasting LDL-C levels were approximately maximal from Week 3 onwards (Figure 3A). Compared with the bococizumab 50-mg and 100-mg arms, the bococizumab 150-mg arm showed stable LDL-C reduction during the treatment period. In the open-label ezetimibe arm, the unadjusted mean percent change (standard deviation) from baseline in fasting LDL-C was −18.6% (16.0) at Week 12 and −20.6% (18.7) at Week 16 (Figure S1A).

Figure 2.

(A,B) Primary analysis: adjusted mean percent changes in LDL-C in the 2 study populations at Weeks 12 and 16. *P<0.001 vs. placebo. CI, confidence interval; LDL-C, low-density lipoprotein cholesterol.

Figure 3.

(A,B) Time course of unadjusted mean percent changes in LDL-C in the 2 study populations. LDL-C, low-density lipoprotein cholesterol; SD, standard deviation.

All subjects receiving bococizumab (100%) achieved LDL-C <100 mg/dL during the treatment period, with all except 1 subject (in the 50-mg arm) also achieving LDL-C <70 mg/dL (Table S2). During the treatment period, an LDL-C level <25 mg/dL was observed in 12.0%, 66.7%, and 58.3% of subjects in the bococizumab 50-mg, 100-mg, and 150-mg arms, respectively.

Bococizumab treatment also caused significant changes in lipid parameters other than LDL-C at both Week 12 and Week 16 (Tables 2,S1). Unadjusted mean percent changes at both time points showed similar patterns to the adjusted values for these parameters (Figure S2A).

Table 2. Adjusted Mean Percent Changes in the Lipid Profile of Hypercholesterolemic Atorvastatin-Treated Japanese Study Subjects Following Treatment

Variable*	Week 12				Week 16
	Placebo (n=26)	Bococizumab			Placebo (n=25)	Bococizumab
	Placebo (n=26)	50 mg (n=25)	100 mg (n=23)	150 mg (n=22)	Placebo (n=25)	50 mg (n=24)	100 mg (n=23)	150 mg (n=22)
LDL-C
Adjusted percent change	−5.18 (−10.45, −0.10)	−55.01^a (−60.39, −49.64)	−71.93^a (−77.55, −66.31)	−76.71^a (−82.35, −71.07)	−11.77 (−16.78, −6.75)	−54.06^a (−59.17, −48.95)	−68.03^a (−73.31, −62.75)	−73.15^a (−78.49, −67.81)
Placebo-adjusted adjusted percent change	–	−49.84 (−57.34, −42.34)	−66.75 (−74.52, −58.99)	−71.53 (−79.28, −63.78)	–	−42.29 (−49.42, −35.17)	−56.26 (−63.60, −48.92)	−61.38 (−68.73, −54.04)
TC
Adjusted percent change	−4.70 (−8.50, −0.91)	−36.34^a (−40.17, −32.52)	−44.58^a (−48.56, −40.61)	−50.59^a (−54.62, −46.56)	−8.95 (−12.71, −5.19)	−34.32^a (−38.11, −30.53)	−42.44^a (−46.34, −38.54)	−47.71^a (−51.69, −43.73)
HDL-C
Adjusted percent change	−6.64 (−10.14, −3.14)	6.75^a (3.19, 10.32)	9.91^a (6.20, 13.62)	6.65^a (2.86, 10.44)	−3.99 (−8.60, 0.62)	6.02^b (1.32, 10.72)	12.26^a (7.44, 17.09)	5.58^b (0.67, 10.49)
TG
Adjusted percent change	4.03 (−6.58, 14.63)	−21.95^a (−32.84, −11.05)	−17.08^b (−28.54, −5.63)	−27.01^a (−38.57, −15.44)	−3.22 (−15.39, 8.95)	−13.28 (−25.77, −0.79)	−27.57^b (−40.49, −14.66)	−18.22^c (−31.23, −5.20)
Non-HDL-C
Adjusted percent change	−4.09 (−8.94, 0.76)	−50.86^a (−55.78, −45.95)	−63.29^a (−68.39, −58.19)	−70.96^a (−76.15, −65.77)	−10.42 (−15.15, −5.68)	−48.41^a (−53.21, −43.61)	−61.27^a (−66.20, −56.34)	−66.77^a (−71.82, −61.73)
Lp(a)
Adjusted percent change	−19.26 (−33.49, −5.02)	−46.70^b (−61.23, −32.18)	−59.85^a (−74.96, −44.74)	−46.42^b (−61.76, −31.08)	−23.44 (−30.64, −16.24)	−41.83^a (−49.18, −34.49)	−60.23^a (−67.80, −52.66)	−60.29^a (−67.98, −52.59)

*Values are presented as mean (95% confidence interval). ^aP<0.001 vs. placebo; ^bP<0.01 vs. placebo; ^cP<0.05 vs. placebo. Abbreviations as in Table 1.

Treatment-Naive Subjects Significant dose-dependent reductions in fasting LDL-C levels were observed in all bococizumab treatment arms compared with placebo at both Week 12 and Week 16 (adjusted mean percent reductions of 47.7–66.8%; P<0.001 for all 3 bococizumab arms at both time points; Figure 2B). The adjusted mean absolute reductions in fasting LDL-C levels observed in the bococizumab arms were also dose-dependent and significant compared with placebo at Week 12 and Week 16 (P<0.001 for all 3 arms at both time points; Table S3). Dose-dependent reductions in fasting LDL-C levels were approximately maximal from Week 3 onwards (Figure 3B). Compared with the bococizumab 50-mg arm, stable LDL-C reductions were observed in the bococizumab 100-mg and 150-mg arms.

With the exception of 2 subjects in the 50-mg arm, all subjects receiving bococizumab achieved LDL-C <100 mg/dL during the treatment period, with 64.9%, 92.0%, and 87.5% in the 50-mg, 100-mg, and 150-mg arms, respectively, also achieving LDL-C <70 mg/dL (Table S2). During the treatment period, an LDL-C level <25 mg/dL was observed in 20.8% of subjects in the bococizumab 150-mg arm, but was not observed in the other treatment arms.

Bococizumab treatment also caused significant dose-dependent changes in lipid parameters other than LDL-C at both Week 12 and Week 16 (Tables 3,S3). Unadjusted mean percent changes at both time points showed similar patterns to the adjusted values for these parameters (Figure S2B).

Table 3. Adjusted Mean Percent Changes in the Lipid Profile of Hypercholesterolemic Treatment-Naive Japanese Study Subjects Following Treatment

Variable*	Week 12				Week 16
	Placebo (n=23)	Bococizumab			Placebo (n=23)	Bococizumab
	Placebo (n=23)	50 mg (n=23)	100 mg (n=22)	150 mg (n=24)	Placebo (n=23)	50 mg (n=23)	100 mg (n=22)	150 mg (n=23)
LDL-C
Adjusted percent change	−1.62 (−7.25, 4.01)	−49.15^a (−54.73, −43.57)	−64.24^a (−69.85, −58.64)	−65.88^a (−71.37, −60.40)	−0.13 (−6.12, 5.87)	−47.70^a (−53.64, −41.77)	−63.47^a (−69.42, −57.52)	−66.82^a (−72.69, −60.94)
Placebo-adjusted adjusted percent change	–	−47.53 (−55.51, −39.55)	−62.62 (−70.56, −54.69)	−64.27 (−72.13, −56.41)	–	−47.58 (−56.07, −39.09)	−63.35 (−71.78, −54.92)	−66.69 (−75.09, −58.30)
TC
Adjusted percent change	−3.92 (−8.15, 0.32)	−34.05^a (−38.25, −29.86)	−43.44^a (−47.69, −39.20)	−45.29^a (−49.43, −41.14)	−1.72 (−5.78, 2.35)	−32.56^a (−36.58, −28.55)	−41.74^a (−45.79, −37.68)	−45.66^a (−49.66, −41.67)
HDL-C
Adjusted percent change	−7.04 (−11.13, −2.96)	−2.93 (−7.02, 1.15)	2.52^a (−1.62, 6.66)	2.78^a (−1.23, 6.79)	−7.44 (−11.45, −3.43)	1.59^b (−2.43, 5.61)	1.10^b (−2.97, 5.18)	1.46^b (−2.52, 5.44)
TG
Adjusted percent change	−7.43 (−20.96, 6.11)	0.61 (−12.78, 13.99)	−9.94 (−23.57, 3.69)	−7.06 (−20.30, 6.19)	4.96 (−8.22, 18.15)	−8.87 (−21.98, −4.24)	1.58 (−11.80, 14.96)	−10.07^d (−23.17, 3.02)
Non-HDL-C
Adjusted percent change	−2.78 (−7.82, 2.26)	−42.84^a (−47.86, −37.81)	−58.79^a (−63.84, −53.74)	−60.36^a (−65.29, −55.43)	0.58 (−4.54, 5.70)	−42.59^a (−47.68, −37.50)	−56.12^a (−61.22, −51.01)	−60.72^a (−65.76, −55.69)
Lp(a)
Adjusted percent change	−17.69 (−30.21, −5.16)	−41.85^b (−54.23, −29.48)	−36.91^c (−49.50, −24.32)	−44.20^b (−56.47, −31.93)	−15.54 (−26.62, −4.46)	−48.48^a (−59.47, −37.49)	−46.86^a (−58.03, −35.68)	−48.39^a (−59.35, −37.42)

*Values are presented as mean (95% confidence interval). ^aP<0.001 vs. placebo; ^bP<0.01 vs. placebo; ^cP<0.05 vs. placebo; ^dP=0.056 vs. placebo. Abbreviations as in Tables 1,2.

Safety

Atorvastatin-Treated Subjects There were no deaths in this subgroup. Treatment-emergent SAEs occurred in 1 subject in the bococizumab 50-mg arm (cellulitis) and 1 subject in the placebo arm (angina pectoris), both of which were assessed by the investigator as not being related to the study drug (Table 4). All treatment-emergent AEs were of mild or moderate severity, with injection-site erythema, injection-site pruritus, nasopharyngitis, and upper respiratory tract inflammation being the most commonly reported AEs in the bococizumab treatment arms. Injection-site erythema and injection-site pruritus were the most frequent treatment-related AEs, although their severity was mild and no subject discontinued study treatment because of an injection-site AE. There was no difference in the incidence of AEs when comparing subjects who achieved fasting LDL-C levels ≤25 mg/dL with those who did not achieve this level (data not shown). There was no increase in abnormal liver function tests over time and no clinically important changes in any treatment arm. No cases of Hy’s law were observed.

Table 4. Summary of Study Safety Data

Variable	Atorvastatin-treated subjects*					Treatment-naive subjects
	Placebo (n=26)	Bococizumab			Ezetimibe 10 mg (open label; n=22)	Placebo (n=23)	Bococizumab
	Placebo (n=26)	50 mg (n=25)	100 mg (n=24)	150 mg (n=24)	Ezetimibe 10 mg (open label; n=22)	Placebo (n=23)	50 mg (n=25)	100 mg (n=25)	150 mg (n=24)
AE, n (%)	13 (50)	17 (68)	16 (67)	13 (54)	5 (23)	11 (48)	16 (64)	16 (64)	15 (63)
Serious AE, n (%)	1 (4)	1 (4)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)**	1 (4)	0 (0)
Treatment discontinuation because of AE, n (%)	0 (0)	0 (0)	1 (4)	1 (4)	0 (0)	0 (0)	2 (8)	2 (8)	0 (0)
Most frequent (incidence ≥5%) treatment-emergent AEs, n (%)
Injection-site erythema	0 (0)	2 (8)	6 (25)	8 (33)	0 (0)	1 (4)	4 (16)	7 (28)	6 (25)
Injection-site hemorrhage	2 (8)	0 (0)	0 (0)	1 (4)	0 (0)	0 (0)	1 (4)	1 (4)	0 (0)
Injection-site pain	0 (0)	0 (0)	1 (4)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	3 (13)
Injection-site swelling	0 (0)	0 (0)	0 (0)	2 (8)	0 (0)	0 (0)	0 (0)	0 (0)	1 (4)
Injection-site pruritus	0 (0)	2 (8)	4 (17)	7 (29)	0 (0)	0 (0)	4 (16)	6 (24)	5 (21)
Pharyngitis	0 (0)	1 (4)	0 (0)	1 (4)	0 (0)	0 (0)	0 (0)	0 (0)	5 (21)
Nasopharyngitis	5 (19)	5 (20)	1 (4)	1 (4)	1 (5)	4 (17)	3 (12)	4 (16)	2 (8)
Upper respiratory tract inflammation	2 (8)	2 (8)	3 (13)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Headache	0 (0)	1 (4)	2 (8)	0 (0)	0 (0)	2 (9)	1 (4)	0 (0)	1 (4)
Dizziness	0 (0)	0 (0)	0 (0)	2 (8)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Fall	1 (4)	1 (4)	1 (4)	0 (0)	0 (0)	1 (4)	0 (0)	3 (12)	0 (0)
Pruritus	0 (0)	0 (0)	2 (8)	0 (0)	0 (0)	0 (0)	1 (4)	0 (0)	0 (0)
Myalgia	1 (4)	1 (4)	1 (4)	0 (0)	0 (0)	1 (4)	3 (12)	0 (0)	0 (0)
Eczema	0 (0)	1 (4)	0 (0)	0 (0)	1 (5)	1 (4)	3 (12)	0 (0)	0 (0)
Blood ALP increased	0 (0)	1 (4)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	2 (8)
Laboratory test abnormalities, n (%)
CK >2×ULN	5 (19)	0 (0)	3 (13)	3 (13)	1 (5)	1 (4)	0 (0)	5 (20)	1 (4)
AST >3×ULN	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	2 (8)	0 (0)
ALT >3×ULN	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	2 (8)	2 (8)

*Hypercholesterolemic subjects with LDL-C ≥100 mg/dL and not controlled by a stable dose of atorvastatin. **Colon cancer occurred in 1 subject and confirmed as serious after the subject’s withdrawal from the study. AE, adverse event; ALT, alanine aminotransferase; ALP, alkaline phosphatase; AST, aspartate aminotransferase; CK, creatine kinase; LDL-C, low-density lipoprotein cholesterol; ULN, upper limit of the normal range.

A total of 2 of the 121 randomized subjects (2%) discontinued treatment because of AEs considered related to the study drug: 1 receiving bococizumab 100 mg experienced a mild gastrointestinal disorder, and 1 receiving bococizumab 150 mg experienced moderate nausea and dizziness; these events were confirmed to have resolved within 15 days and 3 days after onset, respectively.

Treatment-Naive Subjects There were no deaths in this subgroup. Two SAEs occurred in 2 subjects, both of which were considered unrelated to study drug: 1 subject in the bococizumab 100-mg arm experienced a severe SAE of spinal compression fracture, and 1 subject in the bococizumab 50-mg arm was diagnosed with colon cancer in the follow-up period, with this event being confirmed as serious after withdrawal from the study (Table 4). All other treatment-emergent AEs had a maximum severity of mild or moderate, with the most commonly reported being injection-site erythema, injection-site pruritus, nasopharyngitis, and pharyngitis. There was no difference in the incidence of AEs when comparing subjects who achieved fasting LDL-C levels ≤25 mg/dL with those who did not achieve this level (data not shown). Four subjects displayed elevated aspartate aminotransferase and/or alanine aminotransferase activity (>3×the upper limit of the normal range [ULN]), all of whom had mild hepatic function abnormalities at baseline. No cases of Hy’s law were observed.

A total of 3 of the 97 randomized subjects (3%) discontinued study treatment because of AEs considered related to the study drug: 1 receiving bococizumab 50 mg experienced mild injection-site erythema (the same subject who subsequently developed the case of colon cancer described above); 1 receiving bococizumab 50 mg experienced moderate myalgia; and 1 receiving bococizumab 100 mg displayed moderately abnormal hepatic function.

Immunogenicity Overall, 74 (50.3%) of the 147 subjects receiving bococizumab were ADA-positive. A dose-dependent relationship was not observed and there was no apparent difference in the incidence of ADAs between the 2 study populations. In general, there were no apparent differences in LDL-C-lowering efficacy between ADA-positive and ADA-negative subjects across all dosing groups in both study populations (data not shown). However, the onset of ADA in 1 subject (treatment-naive and receiving bococizumab 50 mg) occurred at the same time as a reduction in plasma concentration of bococizumab and attenuation of the LDL-C-lowering response. The incidence of injection-site reactions was higher in ADA-positive subjects than in ADA-negative subjects (40.5% vs. 13.7%). Four subjects who had abnormal hepatic aminotransferase laboratory results (>3×ULN) were ADA-positive.

Discussion

The results from this Phase 2 dose-ranging study in atorvastatin-treated and treatment-naive hypercholesterolemic Japanese subjects demonstrated that bococizumab significantly reduced fasting LDL-C in a dose-dependent manner at both Week 12 and Week 16. The extent of the LDL-C reduction observed was approximately 55–75% of the baseline value in atorvastatin-treated subjects and approximately 50–65% of the baseline value in treatment-naive subjects. The extent of LDL-C reduction we observed in the atorvastatin-treated study population was therefore approximately 5–15% greater than the reduction observed in the treatment-naive subjects. As well as causing a reduction in serum LDL-C levels, treatment with statins has been previously demonstrated to upregulate PCSK9 levels⁹^,¹⁰ and therefore modulation of PCSK9 by background therapy may have influenced the extent of LDL-C reduction observed in the atorvastatin-treated subjects.

The LDL-C-lowering effect observed in our study is greater than the effect observed in a pooled analysis of the 6 lipid-lowering trials within the SPIRE (Studies of PCSK9 Inhibition and the Reduction of vascular Events) Phase 3 clinical development program for bococizumab, and is also greater than the effect observed in a combined analysis of the 2 SPIRE CV outcomes trials.¹¹^–¹³ The pooled analysis of 6 lipid-lowering trials showed a placebo-adjusted reduction in LDL-C from baseline of 55% at Week 12 in patients receiving bococizumab 150 mg q14D, and the combined analysis of the CV outcomes trials showed a placebo-adjusted reduction of 59% at Week 14 in patients receiving bococizumab 150 mg q14D. In contrast, the atorvastatin-treated and treatment-naive subjects who received bococizumab 150 mg q14D in our study displayed reductions of 72% and 64%, respectively, at Week 12. The LDL-C-lowering effect observed in our study was also larger than that observed in a global Phase 2b study of bococizumab in statin-treated subjects in whom mean baseline LDL-C levels ranged from 105 to 119 mg/dL.⁷ The global study reported mean percent LDL-C reductions from baseline of approximately 34–52% at Week 12, although it did not include Japanese participants. The greater LDL-C-lowering effect observed in the current study compared with the pooled analysis of the 6 lipid-lowering trials, combined analysis of the SPIRE CV outcomes trials, and the global Phase 2b study is likely related to the smaller Japanese body size, with weight and body mass index being significant predictors of LDL response in population PK/PD modeling.⁸ The LDL-C-lowering effects observed in our study were of a similar magnitude to the results from a recent meta-analysis of 24 Phase 2 and Phase 3 randomized controlled trials investigating the anti-PCSK9 monoclonal antibodies evolocumab and alirocumab (n=10,159), which reported a mean reduction in LDL-C of 47.49% (95% confidence interval: 25.35–69.64%) vs. treatment with no anti-PCSK9 antibody.¹⁴ The LDL-C-lowering effect observed in our study was also similar to that observed in Japanese trials of evolocumab and alirocumab.¹⁵^–¹⁸ Overall, however, direct comparisons between the current study of bococizumab and the studies described above should be made with caution because the study populations, their baseline characteristics, and the study designs and dosing regimens show considerable differences. Nevertheless, our results add to the growing body of evidence that anti-PCSK9 monoclonal antibody therapies are able to substantially reduce fasting LDL-C in Japanese patients who do not meet guideline-advocated LDL-C targets despite ongoing statin treatment.¹⁹

As well as desirable reductions in fasting LDL-C, the serum levels of other lipid parameters were significantly modified by treatment with bococizumab in a dose-dependent manner; most notably, there were significant reductions in Lp(a). The significance of these changes in other lipid parameters in terms of additional potential cardioprotective effects is difficult to assess and may only become clearer once data from long-term CV outcomes trials have been generated, although disentangling the individual effects contributed by each lipid parameter from the overall changes in such a large number of lipids may be challenging.

In general, the safety profile of bococizumab in our study was similar to that observed in the 6 SPIRE lipid-lowering trials, 2 SPIRE CV outcomes trials, and the global Phase 2b study,⁷^,¹²^,¹³ with the severity of most treatment-emergent AEs being mild. The reduction in LDL-C was substantial and we observed no increase in the incidence of AEs in those who displayed a profoundly reduced level of fasting LDL-C (≤25 mg/dL). We observed a dose-dependent relationship between bococizumab treatment and the incidence of injection-site reactions, which were the most common AE observed. However, these reactions were generally mild in severity and infrequently led to discontinuation of study treatment.

Approximately 50% of all bococizumab-treated subjects were ADA-positive at some point during the study, but there was no dose-dependent relationship with bococizumab treatment and no difference in AEs between ADA-positive and ADA-negative subjects, except for injection-site reactions. There was also no apparent difference in mean LDL-C reduction between ADA-positive and ADA-negative subjects. However, 1 subject who developed ADAs displayed a concurrent reduction in LDL-C-lowering effect from bococizumab treatment, suggesting that a potential neutralizing effect of ADAs on the efficacy of the drug could not be ruled out in this subject. This observation within a single subject in our study is especially interesting given that the pooled analysis of 6 SPIRE lipid-lowering trials reported ADA titer-dependent reductions in bococizumab’s LDL-C-lowering efficacy after the Week 12 timepoint.¹² Overall, bococizumab was well tolerated in both study populations in our study and our data suggested no previously unknown risks from the use of this agent in these subjects. However, on November 1, 2016, Pfizer Inc. announced the discontinuation of the global clinical development program for bococizumab.²⁰ This decision was based on the totality of clinical information for bococizumab, as well as the evolving treatment and market landscape for lipid-lowering agents. Pfizer observed an unanticipated attenuation of the LDL-C-lowering effect over time, as well as a higher level of immunogenicity and higher rate of injection-site reactions than shown with the other agents in this class. Effective and durable lowering of LDL-C is a necessary prerequisite to reduce the occurrence of CV events such as heart attacks and stroke in the long term.

Despite the sponsor’s decision to discontinue the clinical development of bococizumab, it is our hope that the data generated from this Phase 2 study in Japanese subjects will help to inform the wider, ongoing discussion regarding the role of PCSK9 inhibitors within treatment algorithms for the management of hypercholesterolemia. Even if development of bococizumab will not be taken further at this point, the importance of continuing to expand and refine our understanding and clinical armamentarium with regard to atherosclerotic CV disease is difficult to overstate; it remains the most frequent cause of death in the world.²¹ Our data describe the effect of PCSK9 inhibition upon a wide range of lipid parameters in 2 distinct Japanese patient populations, and we hope this information is useful to those planning studies and/or evaluating current and future lipid-lowering therapies. Our 16-week study in an exclusively Japanese population successfully met its primary endpoints and further analysis of the secondary endpoints of this study is ongoing. Pfizer has publically committed to publishing data from the clinical studies of bococizumab despite their decision to discontinue the drug’s development because this disclosure will allow the clinical and research communities to independently assess what these data signify with regard to the much broader topics of PCSK9 inhibitors, lipid-lowering agents, and monoclonal antibody therapies in general.

Conclusions

Treatment of hypercholesterolemic Japanese subjects, who were either receiving a stable dose of atorvastatin or were naive to treatment with lipid-lowering agents, with the anti-PCSK9 monoclonal antibody bococizumab led to significant dose-dependent reductions from baseline in fasting LDL-C of approximately 55–75% at Week 12; similar results were obtained at Week 16. Many other serum lipid parameters were also significantly changed in a dose-dependent manner following bococizumab treatment. Bococizumab treatment was well tolerated by both study populations, with injection-site reactions being the most common AE. Unfortunately, the program was discontinued based primarily on its long-term clinical profile.

Acknowledgments

The authors thank the other principal investigators (Drs Kazuyuki Mizuyama, Shintaro Yano, Harunori Oda, Hyeteok Kim, Koichi Nakamura, and Arihiro Kiyosue), the participating subjects, and the Pfizer Operations Group for their assistance in conducting this study, which was funded by Pfizer Japan Inc. Editorial support was provided by David Wateridge, PhD, of Engage Scientific and was funded by Pfizer.

Conflict of Interest Statement

K.Y. reports research funding from Astellas; departmental sponsorship by MSD; subsidies/donations from Astellas, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Sumitomo Dainippon Pharma, Eli Lilly, Kyowa Hakko Kirin, Mochida Pharmaceutical, MSD, Ono Pharmaceutical, Pfizer, Shionogi, Taisho Toyama Pharmaceutical, Takeda, Mitsubishi Tanabe Pharma, Teijin Pharma, and Toyama Kagaku Kogyo; and honoraria from Astellas, AstraZeneca, Boehringer Ingelheim, Daiichi Sankyo, Sumitomo Dainippon Pharma, Eisai, Kowa, Kowa Pharmaceutical, Kyowa Hakko Kirin, Mochida Pharmaceutical, MSD, Ono Pharmaceutical, Pfizer, Sanofi, Sanwa Kagaku Kenkyusho, Shionogi, Taisho Toyama Pharmaceutical, Takeda, and Mitsubishi Tanabe Pharma. T.T. reports research grants and honoraria from ASKA Pharmaceutical, Astellas, Bayer, Daiichi Sankyo, Kissei, MSD, and Takeda; research grants from Kowa, Mochida Pharmaceutical, and Shionogi; and honoraria from Amgen, Kobayashi Pharmaceutical, Pfizer, and Sanofi. S.K., O.M., and H.S. all report no conflict of interest. K.I., J.T., N.M., and S.C. are employees of Pfizer.

Funding

This study was funded by Pfizer Japan Inc.

Supplementary Files

Supplementary File 1

Supplementary Methods

Supplementary Results

Figure S1. Unadjusted mean percent changes in low-density lipoprotein cholesterol (LDL-C) in the 2 study populations at weeks 12 and 16.

Figure S2. Unadjusted mean percent changes in serum lipids in hypercholesterolemic atorvastatin-treated (A) and treatment-naive (B) subjects.

Table S1. Effect of treatment on the lipid profile of hypercholesterolemic atorvastatin-treated Japanese study subjects

Table S2. Proportion of hypercholesterolemic subjects achieving defined fasting LDL-C levels during the treatment period of the study

Table S3. Effect of treatment on the lipid profile of hypercholesterolemic treatment-naive Japanese study subjects

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-16-1310

References

1. Okamura T. Dyslipidemia and cardiovascular disease: A series of epidemiologic studies in Japanese populations. J Epidemiol 2010; 20: 259–265.
2. Teramoto T, Sasaki J, Ishibashi S, Birou S, Daida H, Dohi S, et al. Absolute risk of cardiovascular disease and lipid management targets. J Atheroscler Thromb 2013; 20: 689–697.
3. Cholesterol Treatment Trialists’ (CTT) Collaboration, Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: A meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet 2010; 376: 1670–1681.
4. Liang H, Chaparro-Riggers J, Strop P, Geng T, Sutton JE, Tsai D, et al. Proprotein convertase substilisin/kexin type 9 antagonism reduces low-density lipoprotein cholesterol in statin-treated hypercholesterolemic nonhuman primates. J Pharmacol Exp Ther 2012; 340: 228–236.
5. Gumbiner B, Udata C, Joh T, Liang H, Wan H, Shelton DL, et al. The effects of single dose administration of RN316 (PF-04950615), a humanized IgG2a monoclonal antibody binding proprotein convertase subtilisin kexin type 9, in hypercholesterolemic subjects treated with and without atorvastatin (abstract). Circulation 2012; 126: A13322.
6. Gumbiner B, Joh T, Udata C, Forgues P, Baum CM, Garzone PD. Effects of 12 weeks of treatment with RN316 (PF-04950615), a humanized IgG2 delta a monoclonal antibody binding proprotein convertase subtilisin kexin type 9, in hypercholesterolemic subjects on high and maximal dose statins (abstract). Circulation 2012; 126: 2782.
7. Ballantyne CM, Neutel J, Cropp A, Duggan W, Wang EQ, Plowchalk D, et al. Results of bococizumab, a monoclonal antibody against proprotein convertase subtilisin/kexin type 9, from a randomized, placebo-controlled, dose-ranging study in statin-treated subjects with hypercholesterolemia. Am J Cardiol 2015; 115: 1212–1221.
8. Wang EQ, Plowchalk D, Gibiansky L, Sweeney K, Kaila N. Population pharmacokinetic and pharmacodynamic modeling of bococizumab (RN316/PF-04950615) in hypercholesterolemic subjects (abstract). Atherosclerosis 2014; 235: e262.
9. Careskey HE, Davis RA, Alborn WE, Troutt JS, Cao G, Konrad RJ. Atorvastatin increases human serum levels of proprotein convertase subtilisin/kexin type 9. J Lipid Res 2008; 49: 394–398.
10. Welder G, Zineh I, Pacanowski MA, Troutt JS, Cao G, Konrad RJ. High-dose atorvastatin causes a rapid sustained increase in human serum PCSK9 and disrupts its correlation with LDL cholesterol. J Lipid Res 2010; 51: 2714–2721.
11. Ridker PM, Amarenco P, Brunell R, Glynn RJ, Jukema JW, Kastelein JJ, et al. Evaluating bococizumab, a monoclonal antibody to PCSK9, on lipid levels and clinical events in broad patient groups with and without prior cardiovascular events: Rationale and design of the Studies of PCSK9 Inhibition and the Reduction of vascular Events (SPIRE) Lipid Lowering and SPIRE Cardiovascular Outcomes Trials. Am Heart J 2016; 178: 135–144.
12. Ridker PM, Tardif J-C, Amarenco P, Duggan W, Glynn RJ, Jukema JW, et al. Lipid-reduction variability and antidrug-antibody formation with bococizumab. N Engl J Med 2017; 376: 1517–1526.
13. Ridker PM, Revkin J, Amarenco P, Brunell R, Curto M, Civeira F, et al. Cardiovascular efficacy and safety of bococizumab in high-risk patients. N Engl J Med 2017; 376: 1527–1539.
14. Navarese EP, Kolodziejczak M, Schulze V, Gurbel PA, Tantry U, Lin Y, et al. Effects of proprotein convertase subtilisin/kexin type 9 antibodies in adults with hypercholesterolemia: A systematic review and meta-analysis. Ann Intern Med 2015; 163: 40–51.
15. Hirayama A, Honarpour N, Yoshida M, Yamashita S, Huang F, Wasserman SM, et al. Effects of evolocumab (AMG 145), a monoclonal antibody to PCSK9, in hypercholesterolemic, statin-treated Japanese patients at high cardiovascular risk: Primary results from the Phase 2 YUKAWA study. Circ J 2014; 78: 1073–1082.
16. Kiyosue A, Honarpour N, Kurtz C, Xue A, Wasserman SM, Hirayama A. A Phase 3 study of evolocumab (AMG 145) in statin-treated Japanese patients at high cardiovascular risk. Am J Cardiol 2016; 117: 40–47.
17. Teramoto T, Kobayashi M, Uno K, Takagi Y, Matsuoka O, Sugimoto M, et al. Efficacy and safety of alirocumab in Japanese subjects (Phase 1 and 2 studies). Am J Cardiol 2016; 118: 56–63.
18. Teramoto T, Kobayashi M, Tasaki H, Yagyu H, Higashikata T, Takagi Y, et al. Efficacy and safety of alirocumab in Japanese patients with heterozygous familial hypercholesterolemia or at high cardiovascular risk with hypercholesterolemia not adequately controlled with statins: ODYSSEY JAPAN randomized controlled trial. Circ J 2016; 80: 1980–1987.
19. Teramoto T, Kashiwagi A, Ishibashi S, Daida H. Cross-sectional survey to assess the status of lipid management in high-risk patients with dyslipidemia: Clinical impact of combination therapy with ezetimibe. Curr Ther Res Clin Exp 2012; 73: 1–15.
20. Pfizer Inc. Pfizer Discontinues Global Development of Bococizumab, Its Investigational PCSK9 Inhibitor [press release]; November 1, 2016. http://www.pfizer.com/news/press-release/press-release-detail/pfizer_discontinues_global_development_of_bococizumab_its_investigational_pcsk9_inhibitor (accessed April 5, 2017).
21. World Health Organization. The top 10 causes of death. http://www.who.int/mediacentre/factsheets/fs310/en/ (accessed April 5, 2017).

責任著者(Corresponding author)

J-STAGEへの登録はこちら（無料）