Author + information
- Received April 4, 2013
- Revision received July 19, 2013
- Accepted July 31, 2013
- Published online December 10, 2013.
- Gregory S. Thomas, MD, MPH∗,†∗ (, )
- William C. Cromwell, MD‡,
- Shariq Ali, PhD§,
- Wai Chin, PhD§,
- JoAnn D. Flaim, PhD‖ and
- Michael Davidson, MD¶
- ∗Long Beach Memorial Medical Center, Long Beach, California
- †University of California, Irvine, Irvine, California
- ‡Lipoprotein & Metabolic Disorders Institute, Raleigh, North Carolina
- §Genzyme Corporation, Cambridge, Massachusetts
- ‖Isis Pharmaceuticals, Carlsbad, California
- ¶University of Chicago, Chicago, Illinois
- ↵∗Reprint requests and correspondence:
Dr. Gregory S. Thomas, MemorialCare Heart & Vascular Institute, Long Beach Memorial Medical Center, 2801 Atlantic Avenue, Long Beach, California 90806.
Objectives This study sought to examine the efficacy and safety of mipomersen for reducing atherogenic lipids and lipoproteins in patients with hypercholesterolemia.
Background Many patients on lipid-lowering therapies remain unable to achieve target low-density lipoprotein (LDL) cholesterol levels. Mipomersen, an antisense oligonucleotide inhibitor of apolipoprotein B, reduces LDL cholesterol and atherogenic lipoproteins.
Methods This randomized, double-blind, multicenter study enrolled 158 patients with baseline LDL cholesterol levels ≥100 mg/dl with, or at high risk for, coronary heart disease who were receiving maximally tolerated lipid-lowering therapy. Patients received weekly subcutaneous mipomersen 200 mg (n = 105) or placebo (n = 52) for 26 weeks, with a 24-week follow-up period. Randomization was stratified by type 2 diabetes status.
Results Sixty mipomersen and 44 placebo patients completed treatment. Mean baseline LDL cholesterol levels were 122.7 and 122.6 mg/dl in the placebo and mipomersen patients, respectively. Mipomersen reduced LDL cholesterol by −36.9% compared with placebo at −4.5% (p < 0.001). Target LDL cholesterol <100 mg/dl was attained in 76% of mipomersen and 38% of placebo patients. Mipomersen also significantly reduced apolipoprotein B (−38%) and lipoprotein(a) (−24%) (p < 0.001). Common adverse events included injection site reactions (78% with mipomersen, 31% with placebo) and flu-like symptoms (34% with mipomersen, 21% with placebo). Elevations in transaminases and liver fat also occurred in some patients, and these levels returned toward baseline after treatment cessation.
Conclusions Mipomersen significantly reduced LDL cholesterol, apolipoprotein B, and lipoprotein(a) in patients with hypercholesterolemia with, or at risk for, coronary heart disease not controlled by existing therapies. (Safety and Efficacy of Mipomersen [ISIS 301012] as Add-On Therapy in High Risk Hypercholesterolemic Patients; NCT00770146)
- antisense oligonucleotides
- apolipoprotein B
- cholesterol inhibitors
- hypolipidemic agents
- lipid-regulating agents
Low-density lipoprotein (LDL) is key in the pathogenesis of coronary heart disease (CHD). LDL particles enter the arterial wall through a gradient-driven process. Once inside the intima, LDL particles that bind to arterial wall proteoglycans are retained, oxidized, and subsequently taken up by macrophages to form foam cells (1). LDL particle–lowering agents such as statins significantly reduce CHD risk (2,3). National Cholesterol Education Program Adult Treatment Panel guidelines emphasize the importance of LDL management to reduce CHD risk. Among high-risk patients with known CHD, the guidelines recommend an LDL cholesterol level of <100 mg/dl (3). However, conventional lipid-lowering therapies often result in insufficient LDL cholesterol reductions, even when administered at maximally tolerated doses (4).
Apolipoprotein B (apoB) is an essential component of very-low-density lipoprotein (VLDL), intermediate-density lipoprotein, LDL, and lipoprotein(a) (Lp[a]), with 1 molecule of apoB present in each lipoprotein particle. ApoB is constitutively expressed in the liver. The consequences of pharmacologic inhibition of apoB synthesis are unknown and include the potential of hepatic compensation via increased beta oxidation of hepatic lipid, as well as steatosis. Mipomersen, an antisense oligonucleotide, decreases apoB synthesis by inhibition of messenger ribonucleic acid translation (Fig. 1) (5–7). Mipomersen has significantly reduced LDL, apoB, and Lp(a) in patients with homozygous familial hypercholesterolemia (FH) and moderate or severe heterozygous FH (8–10). We evaluated the safety and efficacy of mipomersen compared with placebo in patients with hypercholesterolemia with, or at high risk for, CHD already receiving a maximally tolerated lipid-lowering regimen. This is the first phase 3 evaluation of mipomersen in patients without FH.
This prospective, randomized, double-blind, placebo-controlled study was conducted at 62 U.S. centers between November 2008 and October 2010. After providing informed consent and undergoing screening, eligible patients were randomized (2:1) to mipomersen 200 mg or placebo. Randomization was stratified so that a minimum number of patients (40%) would have type 2 diabetes mellitus (T2DM). Medication was administered as a single, subcutaneous injection once weekly for 26 weeks, allowing assessment at steady-state levels of mipomersen given its half-life of approximately 31 days (7). Patients then entered a 24-week safety follow-up. This trial (NCT00770146) was approved by all ethics boards and conducted according to Good Clinical Practice and International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use guidelines.
Men and nonpregnant, nonlactating women age ≥18 years with hypercholesterolemia (fasting LDL cholesterol ≥100 mg/dl, triglyceride <200 mg/dl) with, or at high risk for, CHD (per National Cholesterol Education Program Adult Treatment Panel III guidelines) were eligible (3). At screening, all patients were at stable weights, on low-fat diets, and receiving lipid-lowering regimens that included a maximally tolerated statin dose. Major exclusion criteria included significant cardiovascular or cerebrovascular events within 24 weeks of screening, congestive heart failure, type 1 diabetes, uncontrolled hypertension, any disorder known to predispose to secondary hyperlipidemia, or a history of renal or hepatic disease. Patients were not permitted to alter their lipid-lowering regimens for 28 weeks.
Patients were evaluated at baseline and every other week for the first 9 weeks, every 4 to 5 weeks for the remainder of treatment, and 4 times during follow-up. Laboratory assessments and statistical analysis were similar to a previously published trial (8), as described in Online Appendix A. The primary outcome was percent reduction in LDL cholesterol from baseline to the primary efficacy timepoint, defined as the post-baseline visit closest to 14 days after the last dose of medication (week 28). Additional efficacy outcomes included percent changes in apoB, total cholesterol, non–high-density lipoprotein (HDL) cholesterol, triglyceride, Lp(a), VLDL cholesterol, LDL/HDL ratio, apolipoprotein A-I, and HDL cholesterol. An exploratory analysis of lipoprotein particles occurred.
Online Figure 1 displays patient flow. At baseline, among patients who received the study treatment (n = 157), 72% had metabolic syndrome (11), 56% had T2DM, and 52% had CHD or other atherosclerotic disease. All patients were receiving lipid-lowering medications, including statins; 43% were receiving the maximal U.S. Food and Drug Administration–approved statin dose. Patients were typically middle aged, white, and overweight, with a mean body mass index of 30.4 ± 4.6 kg/m2 (Table 1). Groups were well balanced in demographics and baseline characteristics; there were no clear differences between patients with and without T2DM.
Mean LDL cholesterol was elevated at baseline: 122.6 ± 31.7 mg/dl in the mipomersen group and 122.7 ± 38.6 mg/dl in the placebo group (Table 2). At the primary efficacy timepoint, mipomersen-treated patients experienced significantly greater reductions from baseline in mean LDL cholesterol (−36.9%) than placebo-treated patients (−4.5%) (p < 0.001) (Fig. 2). Moreover, 76% of mipomersen versus 38% of placebo patients attained LDL cholesterol <100 mg/dl at the primary efficacy timepoint, while 51% of mipomersen versus 8% of placebo attained LDL cholesterol <70 mg/dl. Substantial reductions in mean LDL cholesterol (−17%) were seen by week 5, with near maximal effects observed by week 17, consistent with a half-life of approximately 31 days (7); after treatment completion (week 26), LDL cholesterol gradually increased and returned to baseline by week 50 (Fig. 3).
The effect of mipomersen was greater in women and patients older than 59 years (mean/median age) compared with men and patients younger than 59 years, whose lipid lowering was still clinically meaningful and statistically significant. The mean percent changes in LDL cholesterol were −32.7 ± 25% and −41.2 ± 28.3% for male and female mipomersen patients, respectively, compared with placebo at −8.6 ± 26.6% and −1.1 ± 19.7%. The mean percent changes in LDL cholesterol were −29.8 ± 30.3% and −41.8 ± 23.3% for mipomersen patients younger and older than 59 years, respectively, compared with −7.4 ± −25.2% and −1.9 ± 23.5% for placebo. Baseline LDL cholesterol and race did not influence treatment effect.
Reductions from baseline to the primary efficacy timepoint in apoB, non-HDL cholesterol, and LDL/HDL ratio were similar (about 37%) to those in LDL cholesterol; reductions in total cholesterol, triglyceride, and VLDL cholesterol were slightly smaller (about 25%) (Table 2, Fig. 3). Lp(a), not typically affected by statins, ezetimibe, or bile acid sequestrants (12), was also reduced (26% with mipomersen, 0% with placebo). A small decrease in apolipoprotein A-I was observed with mipomersen. No significant difference was noted between treatment groups in HDL cholesterol or high-sensitivity C-reactive protein. As with LDL cholesterol, all lipids and lipoproteins gradually returned to baseline after treatment cessation. Baseline levels of lipids and lipoproteins and changes in lipids and lipoproteins associated with mipomersen treatment were similar regardless of diabetic status (Table 3).
In an exploratory analysis, lipoprotein particle concentrations were determined using nuclear magnetic resonance spectroscopy. As in previous studies (13,14) greater reductions were noted in the number of small (−542.5 nmol/l) versus large LDL particles (−79.7 nmol/l) in the mipomersen group; no meaningful changes occurred in the placebo group (small, −54.1 nmol/l; large, −22.6 nmol/l).
The most common adverse events (AEs) related to tolerability were injection site reactions (ISRs) and flu-like symptoms (FLS). The incidence of ISRs was greater with mipomersen (78.1%) compared with placebo (30.8%), as was the incidence of FLS (34.3% vs. 21.2%) (Table 4). The most common AEs related to safety were elevated liver enzyme levels and increased liver fat. Alanine aminotransferase (ALT) increased in more mipomersen patients compared with placebo but trended toward baseline after week 26 (Fig. 4). Ten mipomersen patients had ALT levels ≥3 times the upper limit of normal on 2 consecutive measures at least 7 days apart; no concomitant increases in bilirubin or changes in liver synthetic function occurred.
The mean of absolute changes in average liver fat fraction from baseline to end of treatment was 15.9 ± 11.2% for mipomersen (n = 48) versus 0.6 ± 7.1% for placebo (n = 33) (Table 5). The mean liver fat fraction remained unchanged at the end of the week 24 follow-up in the placebo group (1.3 ± 7.7%) and decreased in the mipomersen group (5.1 ± 7.5%). Results were similar when summarized by diabetes status. Post-hoc analyses were performed to explore correlations in liver parameters (Online Figs. 2A to 2D). As expected, in the mipomersen group, there was a strong correlation between the percent changes in apoB and LDL cholesterol levels from baseline to the primary efficacy timepoint (r = 0.95). Weaker associations were observed between percentage change in apoB and change in liver fat fraction (r = −0.52), percent change in apoB and maximal ALT level (r = −0.44), and change in liver fat fraction and maximal ALT level (r = 0.52). Except for the close association of LDL cholesterol versus apoB (r = 0.80), there was no correlation in the placebo group.
Of the 54 patients who discontinued, 45 were in the mipomersen group and 9 in the placebo group. AEs were responsible for 28 (18%) of the discontinuations: 26 (25%) in the mipomersen group (7 because of liver enzyme elevations and other AEs and 7 because of ISRs) and 2 (4%) in the placebo group (Online Appendix B). Discontinuation was approximately 11% of patients in both groups secondary to withdrawal of consent. Serious AEs occurred in 7 (7%) mipomersen and 4 (8%) placebo patients. One death occurred in each group. The placebo patient died during the treatment period from acute myocardial infarction and cardiogenic shock. The mipomersen patient was admitted to the hospital with acute myocardial infarction and pneumonia 149 days after completing treatment with mipomersen; the patient died the following day from acute liver failure. The case was adjudicated by 2 independent hepatologists, who concluded that the cause of death was hepatic failure due to acetaminophen toxicity and was unrelated to mipomersen.
Mipomersen had no adverse effect on renal function (serum creatinine, estimated glomerular filtration rate, blood urea nitrogen), muscle (creatinine kinase), hematology, glucose, or blood pressure, nor were any clinically meaningful safety findings or trends in weight, vital signs, or electrocardiographic parameters noted. T2DM status did not alter the safety profile in these parameters or in relation to ISRs, FLS, or hepatic transaminase elevation or steatosis.
This is the first phase 3 study of mipomersen in non-FH patients with high cardiovascular risk due to prior CHD events and/or concurrent T2DM. Mipomersen 200 mg weekly significantly reduced LDL cholesterol (by 36.9%) compared with placebo (by 4.5%); the onset of action was as early as 5 weeks. Consistent with its mechanism of action, reductions in total cholesterol, non-HDL cholesterol, LDL/HDL ratio, triglyceride, VLDL cholesterol, and Lp(a) were also sizable; these effects were similar in patients with diabetes and greater in women and older patients. The mipomersen dropout rate was 43%; AEs accounted for 25% and included ISRs, FLS, and ALT increases. The placebo dropout rate was 17%; AEs accounted for 4%. AEs affecting tolerability included ISRs and FLS. ISRs did not occur at every injection site or in all patients; 20% of mipomersen patients had no ISRs. Anecdotally, ISRs lessened with pre-dose oral diphenhydramine, icing, or topical lidocaine, and FLS decreased with pre-dose nonsteroidal anti-inflammatory drugs.
Either altering cholesterol secretion from the liver, as would be expected by lowering the synthesis of apoB, or another unspecified effect resulted in a mean change of 40.9 ± 51.4 U/l in ALT from baseline to treatment end. Mean ALT decreased to near baseline at the end of the 24-week treatment follow-up. Hepatic fat content and ALT were correlated (r = 0.52) with decreasing apoB, which was strongly correlated with decreasing LDL (r = 0.95). Mean hepatic fat content decreased to near normal by 24 weeks after treatment. It is not known if liver adaptation with the potential to normalize hepatic fat would occur if mipomersen had been continued longer. Adaptation has been observed in mice receiving murine apoB antisense oligonucleotide, whereby compensatory changes occur in pathways of hepatic lipogenesis and fatty acid oxidation (15). Further to this observation, an interim analysis of the 2-year mipomersen open-label extension trial found hepatic fat content to increase during the first year in some patients but to stabilize or decrease with continued treatment (16).
In the present study, which included patients with diabetes, there was no indication of clinical sequelae associated with increases in ALT levels and/or liver fat content. Liver biopsies obtained from mipomersen-treated subjects in other studies have confirmed steatosis and found minimal inflammation with little to no fibrosis (17). By extension, some but not all patients with familial hypobetalipoproteinemia, a lifelong condition of reduced apoB levels, have steatotic livers. This secondary condition is not associated with insulin resistance (18,19).
In this study, fewer patients completed treatment compared with phase 3 trials of similar designs enrolling patients with FH (8–10). One theory may be a reduced sense of treatment urgency by physicians treating patients without known genetic diseases and with lower baseline LDL cholesterol levels. Additional safety studies in this population and others will be necessary to fully explain high dropout rates and potential opportunities to mitigate the occurrence of AEs. A gradual increase in the dose of mipomersen is being evaluated, for example, in the Study of the Safety and Efficacy of Two Different Regimens of Mipomersen in Patients With Familial Hypercholesterolemia and Inadequately Controlled Low-Density Lipoprotein Cholesterol in patients with heterozygous FH (NCT01475825). At this time, mipomersen is approved for the treatment of homozygous FH. The clinical development plan remains focused on patients with genetically derived hypercholesterolemia, with the greatest therapeutic potential in patients with refractory FH.
Mipomersen represents a first-in-class injectable antisense therapy and provides the opportunity to use novel antisense technology to modulate messenger ribonucleic acid translation without altering deoxyribonucleic acid. Mipomersen, when added to lipid-lowering therapy, significantly decreased LDL cholesterol, apoB, Lp(a), and other atherogenic lipoproteins, potentially providing a new treatment option for patients. The relatively high discontinuation rate attributed to ISRs and FLS should encourage clinicians to focus on managing patient expectations.
The authors thank the investigators (Online Appendix C) and site coordinators for their diligence in data acquisition. Barbara Rinehart of Research Pharmaceutical Services, Inc., contracted by Genzyme, provided writing assistance. Brenda Baker of Isis Pharmaceuticals, Inc., provided critical review of the manuscript.
This study was sponsored by Isis Pharmaceuticals and Genzyme Corporation, a sanofi-aventis company. Dr. Thomas is a consultant for Genzyme, a sanofi-aventis company; and has received research grant support from Genzyme, Regeneron, and sanofi-aventis. Dr. Cromwell is a consultant for Genzyme/Isis, LabCorp, and Health Diagnostic Laboratory; has received research grant support from Isis Pharmaceuticals; and is a speaker for Abbott, Kowa, Merck, and LipoScience. Drs. Ali and Chin are Genzyme employees. Dr. Flaim is an Isis Pharmaceuticals employee. Dr. Davidson is a consultant for Genzyme and Sanofi; and has received research grants from Isis Pharmaceuticals, Genzyme, sanofi-aventis, Regeneron, and Amgen.
- Abbreviations and Acronyms
- adverse event(s)
- alanine aminotransferase
- apolipoprotein B
- coronary heart disease
- familial hypercholesterolemia
- flu-like symptoms
- high-density lipoprotein
- injection site reaction
- low-density lipoprotein
- type 2 diabetes mellitus
- very-low-density lipoprotein
- Received April 4, 2013.
- Revision received July 19, 2013.
- Accepted July 31, 2013.
- American College of Cardiology Foundation
- Tabas I.,
- Williams K.J.,
- Boren J.
- Grundy S.M.,
- Cleeman J.I.,
- Merz C.N.,
- et al.
- Kastelein J.J.,
- Wedel M.K.,
- Baker B.F.,
- et al.
- Raal F.J.,
- Santos R.D.,
- Blom D.J.,
- et al.
- Stein E.A.,
- Dufor R.,
- Gagne C.,
- et al.
- Grundy S.M.,
- Cleeman J.I.,
- Daniels S.R.,
- et al.
- Cromwell W.,
- Dufour R.,
- Gagne C.,
- et al.
- Cromwell W.C.,
- Santos R.D.,
- Blom D.J.,
- et al.
- Lee R.G.,
- Fu W.,
- Graham M.J.,
- et al.
- Duell P.B.,
- Santos R.D.,
- East C.,
- et al.
- Visser M.E.,
- Wagener G.,
- Baker B.F.,
- et al.